The specifications of any microphone preamplifier make interesting reading. There are three most important points to consider – frequency response, distortion and signal-to-noise ratio.
The frequency response of any purely electronic piece of equipment these days should be ruler-flat within the range of human hearing, and to a reasonable safety margin beyond. For a competent designer, this is easy. The Focusrite LE claims -0.4 dB at 10 Hz to -3 dB at 122 kHz at minimum gain. At maximum gain the specification says -2.3 dB at 10 Hz, -3 dB at 67 kHz. So within the accepted normal range of human hearing, you would be hard pressed to find a ruler so flat.
The distortion figure (actually total harmonic distortion plus noise – 'THD+N' – as it is more practical to measure) is at worst 0.003% with a gain of 60 dB, input level -36 dBu and a 20 Hz to 22 kHz filter switched in on the measurement device. This would make the level of the distortion around 90 dB below a 0 dBu output level signal. If you can hear this, then you have supernatural powers and someone should make a TV series about you.
When we come to the noise level, first we need a little understanding of how noise is measured and stated for microphone preamplifiers. Uniquely, the noise figure is given as the 'equivalent input noise', or EIN. To explain this briefly, the noise level at the output is measured, and then the gain is subtracted from that. So if the gain is set to 60 dB, and the output noise level is -64 dBu (meaning 64 dB below 0.775 volts), then the EIN is -124 dBu, which is the case with the Focusrite Octopre LE.
But there is more to it than that. The noise level depends to a large extent on the impedance of the microphone that is connected. Think of it this way – the lower the impedance; the more electrons share the duty of carrying the signal; the more their random vibrations (the noise) will tend to even out. Therefore the lower the impedance, the lower the noise.
Typically a microphone has an impedance of around 200 ohms, which sets a minimum noise level due to random vibrations of electrons that no microphone preamplifier can better. This absolute minimum EIN therefore is around -129 dBu and the only way that can be exceeded is to cool the microphone and preamp in liquid nitrogen (DON'T do that!).
Various manufacturers of course have their methods of 'massaging' their EIN figures to make them look better – using a lower than normal source impedance (or even a dead short); using a weighting filter that mimics the response of the human ear (but isn't applicable at these very low levels); or using an uncommonly high gain, where better EIN figures tend to be found.
Of these 'specification improvement' techniques, Focusrite – like just about every other manufacturer – do quote their EIN at a very high gain. But their impedance is a realistic 150 ohms, and the filter is from 22 Hz to 22 kHz, which is reasonable.
In comparison, the Avalon AD2022 preamp, which is damned expensive and has just two channels, claims an EIN of -126 dB (presumably dBu), unweighted and with a source impedance of 150 ohms. The value-for-money Spirit by Soundcraft Notepad claims -128.5 dBu (150 ohms) but with no mention of weighting. One suspects that this is a weighted figure, which would probably make the unweighted figure around -123.5 dBu.
The odd thing about microphone preamplifiers is how closely manufacturers are pushing the limits of the possible. And how guarded they are about noise figures at typical gain settings of +20 to +30 dB. Still, there are few instances where any properly designed microphone preamplifier should be found wanting in terms of frequency response, distortion and noise, and the Focusrite Octopre LE is no exception.