Yes, headroom can be an important fact (and I have found this to be particularly true in recording consoles, where the input signals contain such vastly different, uncompressed levels). But, relative to this thread, here is something to consider. Decades ago, I was involved in writing the review of a Hi-Fi audio preamp from a well known manufacturer (not a microphone preamp). The magazine that commissioned the review had a policy of letting the gear maker pre-read reviews in case there was anything they wanted to address, or even correct in production models. To create this review, measurements were taken, extensive listening tests were done, and the review was drafted. It mentioned one or two less-than-stellar aspects of the sound of that preamplifier. When the manufacturer read the review they took great exception to our findings and challenged the primary author and me to do a blind A/B/C test. They had only two conditions: any circuits being compared to theirs had to be set within 1/2 dB of playback volume at 1KHZ, and the frequency response of any circuit we wanted to include had to fall within 1/2 dB of their circuit (which measured very flat from very low to very high). My friend and I took up the challenge with great swagger, and we were ready to "set them straight".
Too bad for us. When we met their criteria, no one who heard the tests could distinguish between any of the several circuits we ended up testing on that day. We spent hours and tried various topologies, tube/SS, P-to -P- wiring vs circuit boards; we pretty much covered it all. WHEN THE FREQUENCY RESPONSES AND PLAYBACK VOLUME WERE MATCHED WITHIN 1/2 dB, NO ONE COULD DISTINGUISH THE CIRCUITS. By the way, I don't mean that we EQ'd frequency responses to be within 1/2 dB, I mean that we only used circuits that, by design, were that close in frequency response right out-of-the-box (20-20K).
We then focused on some listening that considered noise/distortion content. What we found was that we had to add so much noise/distortion in order to come up with repeatable differentiations between the units that we knew no one would ever even try to sell such a noisy/distorted circuit.
I have never fully gotten over the impact of participating in these tests. I owned exotic great sounding gear. I had ears that could tell the difference between subtle changes in circuits and systems. I had done many tests previously and the results consistently supported my belief that the sonic differences between circuits were palpable. One last detail: We measured some units that had frequency responses that were only slightly farther off from the test preamplifier than 1/2dB. Guess what, we could tell the difference, and we had preferences.
Nowdays, I STILL buy gear that is highly rated and glowingly reviewed. The difference is that since the day of that test, I realize that at least THE PRIMARY differences in PREAMPLIFIERS (to address the topic of the OP) are locateable in frequency response. Do I have a playback system that I think sounds "wide", and has fantastic dynamics, etc., etc. I sure do. But when it comes to PREAMPLIFIERS, I think the differences are far smaller than amps, and vastly smaller than they are between speakers, microphones, and phoro cartridges. Very careful measurement of FR will tell you a great deal about "Quickness", "Slam", and other audiophile categories of performance.
If you have a modicum of test gear, try the comparison test for yourself. But remember, you only get to have plus-or-minus 1/4 dB in frequency response and playback volume (at 1KHZ) between the examples you are trying to compare. Let me know if you hear a difference. Maybe it's time for me to learn something new - again!
