Comparing mic self-noise specs

S

Samuel Dilworth

Well-known member
Below are the self-noise specifications for two popular shotgun microphones in the terms used by the manufacturers:

Rode NTG3
Equivalent noise level (A-weighted): 13 dBA

Sennheiser MKH 416
Equivalent noise level A-weighted: 13 dB
Equivalent noise level CCIR-weighted (CCIR468-3): 24 dB

Okay, so the comparable specs are identical at 13 dB.

However, in this case the Rode’s sensitivity is a little higher than the Sennheiser’s, so it needs less amplification. Are the figures directly comparable or do we have to account for the different sensitivity?

Here it doesn’t matter so very much (though it still favours the Rode), but often the sensitivity difference between two mics is much greater.

Thanks.
 
The sensitivity between mics is often talked about, but unless the difference is severe (as in some dynamics) in practice the extra gnats whisker on the gain control makes no difference at all. I often wonder if the ONLY time we look at the specs is when we perceive a problem. You stick it in, you turn it up and if the result is a success, we are happy. If we really wanted to compare the 416 and the NTG3, then they'd need to have the same polar pattern, or we'd not be measuring like with like anyway?
 
The difference in sensitivity is nearly negligible (-32dBV vs -30dBV), but either way the 13dB self-noise is a very good rating. Keep in mind that manufacturer specs are not always entirely accurate as they do their bench tests under their own controlled circumstances, and numbers can be fudged a little. But self-noise ratings are still worth paying attention to, and 13dB is a rather quiet mic.

The biggest differences between the NTG-3 and the 416 are in off-axis rejection (the 416 is a little more focused whereas the NTG-3 has a slightly wider pickup) and in frequency response (overall sound).

Of course, the only way to know for sure what any of the numbers mean to a mic is to plug it in and listen to it.
 
I had an early NTG3 in for a review; it was OK I. I mean, professionally OK. I have a 416 that I've had forever; I use it very little these days because I prefer other short-shotgun mics (Schoeps CMIT and Sanken CS3e), but that's just me. As Paul and Alex say (and correct me if I'm misreading you), out in the streets or just about anywhere besides perhaps a recording studio, the noise specs of most professional mics (such at the NTG3 and 416) aren't an issue...especially if you're recording dialog and not silence. :) And the preamps in good professional modern audio recorders can handle most sensitivity differences without creating big noise-floor problems.

So are you trying to decide between these two mics? How do you plan to use them? Docs, fiction narrative? Out in the streets or in more controlled environments? Will your choice be you only "good" mic? Do you have a budget limit?
 
Understood. But let’s take mics with bigger differences:

NTG1
Sensitivity: −36.0 dB
Self-noise: 18 dBA

MKH 60 (a leeetle bit more expensive)
Sensitivity: −28.0 dB
Self-noise: 8 dBA

Is the 18 dBA figure directly comparable to the 8 dBA figure or do we have to add the 8 dB difference in sensitivity too, since the recorder gain has to be set that much higher for this Rode than this Sennheiser? (And then the recorder’s preamp might add its own significant noise at the higher gain … or not).

Jim Feeley said:
So are you trying to decide between these two mics?
Click to expand...
No, just trying to understand the specs. I understand the limited descriptive power of these two narrow specs (self-noise and sensitivity), but I’m still interested in what they say.

In this unrigorous but useful test by Curtis Judd, the NTG3 and MKH 416 appear to have practically identical noise to my ear (go to 2:56 for the NTG3 and a little later for the MKH 416).

https://www.youtube.com/watch?v=bv1xUShuUY0
 
Thinking about this some more, it seems to me that the recorder preamp is unlikely to add significant noise with any of these mics although a dynamic mic would obviously be a different story.

The self-noise versus sensitivity figures remain intriguing. Starting to think the self-noise must somehow account for sensitivity differences, based on looking at specs of various mics I’m vaguely familiar with.
 
Jim Feeley said:
As Paul and Alex say (and correct me if I'm misreading you), out in the streets or just about anywhere besides perhaps a recording studio, the noise specs of most professional mics (such at the NTG3 and 416) aren't an issue...especially if you're recording dialog and not silence.
Click to expand...

Not exactly. Noise specs do matter, to an extent. Example: the Shure VP-88 has an extremely high self-noise of 25dB. It’s great for stereo tracks in higher-SPL situations, but it’s dreadful for trying to record softer ambient beds. Being a stereo mic, it’s a terrible choice for dialog, but for the purpose of this thread it really has to do with the self-noise vs. the levels of the source you’re recording. I see the VP-88 in use often on NBC’s Today with choirs out on the plaza, which fits because the gain will be reduced and the ambient noise floor of NYC is going to obscure any of the mic’s self-noise after that.

A mic with lower self-noise, say 12dB, is going to work just fine in a quiet dialog scene. In practice, the difference between 8dB and 12dB for on-location dialog isn’t going to matter so much. 18dB however...

Samuel Dilworth said:
Thinking about this some more, it seems to me that the recorder preamp is unlikely to add significant noise with any of these mics although a dynamic mic would obviously be a different story.
Click to expand...

18dB of self noise is fairly high. I tend to shy away from anything above 16dB if I need clean recordings of softer sources. Really, 13dB and below are preferable. 8dB is incredibly quiet.

As for recorder pre-amps, it really depends on the recorder. Cheaper recorders have higher self-noise in the pre-amps and therefore need mics with higher output levels, thus requiring less amplification in the pre-amp. The NTG-1 and NTG-2 are notoriously bad choices to pair with something like a Zoom H4n or Tascam DR-60D because those recorders’ pre-amps don’t have as much clean gain. But then, those mics also have a high noise floor of their own.

Output impedance also has a lot to do with it, as higher impedance mics have weaker output signals and need more clean gain in the pre-amp. The NTG-2 has an impedance of 250Ω (phantom-powered) or 350Ω (battery-powered), as opposed to the Audio Technica AT875r that has a much hotter output at 100Ω impedance.

TL;DR - It’s not just sensitivity or self noise. Those both have to be taken into account, along with the mic’s impedance AND the amount of clean gain in the mixer/recorder pre-amp.
 
I'm not very familiar with the NTG1/NTG2. I've never recorded with one, but I have worked with tracks recorded with that mic. Don't recall self noise being a big issue. IIRC, I mostly thought that there was a lot of background noise (ie- how and where the tracks were recorded) and that the mic was a touch harsh but better than other low-cost mics then popular (e.g., the ickypoo ME66/K6). Harsh because Rode couldn't cram $1000 worth of parts into a $200 mic (duh. :happy:) And 18dBA is a non-optimal but still-decent spec for our work (imo).

To your question: For production sound in many uses, I =think= self-noise difference between the NTG1 and MKH60 won't matter all that much if you're running the mic through a high-end Sound Devices or similar-quality recorder such (SD 633, 442, 844, Scorpio; Zaxcom Nova, or whatever). And I think it also won't matter much through a current MixPre and perhaps Zoom F-series, but I don't own those. But then, I'm thinking of my work. Not of being in a studio or something.

I feel in general that Rode makes decent products for their price points, but I'd wonder if their specs play a bit of a marketing role. For Sennheiser's professional mics, I'd think their specs are more informative and mainly designed to clue in potential customers about how they might want to work with the mic. But for production sound, there's so much else that matters.

I have a decent understanding of the specs and how they're generated. Long long ago, I built microphones and other equipment for recording studios and touring musicians (note that I didn't design the equipment; I just built it and I was a technician). We would take each mic we made and and generate a a frequency-response and specs chart for that specific mic. The recording engineers appreciated that, but it didn't play a big roll in deciding if they'd buy the mics or even in how they'd use them (though it didn't hurt). Purchase decisions came from word of mouth, mentions and reviews in magazines such as Recording Engineer/Producer and dB, and hands on experience (loaners were important). So these days I follow the same pattern: what mixers I respect think, some comments from forums like this, and personal experience with rentals or loaners. Don't care about specs much. Oh; I check out polar patterns since terms like hyper and super aren't very specific.... But still, the experiences and thoughts of others and of myself matter more.

I don't follow Curtis Judd, but I do follow Ray Charles who legendarily/supposedly said to a recording engineer something like, "I don't care how many channels you have. How does it sound baby?"

Less snarkily: I'd probably pay more attention to Judd than to Rode's specs. I glanced at Judd's video just now. perhaps his test can help narrow purchase decisions (ie- mainly to eliminate mics from consideration). And nothing against him, because there are so many variables that are hard to account for. You gotta get the mic on location and see how it works for you in the real world.

Specs only go so far. And it helps me to consider a mic as part of a system comprised of the location, the speaker(s)/sound source, the mic, the mixer/recoder, and the ultimate delivery medium.

Several microphone makers have informative sites that I find helpful. Sure, they're marketing but they're positive marketing and not only hype. Two that come to mind are Neumann and. DPA. Here are their respective pages on evaluating self-noise specs:

WHAT IS SELF-NOISE (OR EQUIVALENT NOISE LEVEL)?
https://www.neumann.com/homestudio/en/what-is-self-noise-or-equivalent-noise-level

THE BASICS ABOUT NOISE IN MICS
https://www.dpamicrophones.com/mic-university/the-basics-about-noise-in-mics

Others here will probably add differing opinions (and be ready for the inevitable tangent into the qualities of different studio preamps #Neve1073forever :violent5:).

Sorry to come off as arrogant and dismissive. HTH a little...
 
Last edited:
I don't think I have ever used a mic that gave me a nudge to look upon specs from it's performance. The only mic in the past five years that when I plugged it in, I thought "eh?" was a faulty one - Even the SM7B, a very low output mic has never made me wince by it's noise. This is like the topics we have own camera sensitivity. My favourite video cameras have always been the subject of "terrible low light performance' in various and repetitive topics, but I've alway used them in theatres, on stage, and low light has never been a problem. On paper, they are poor performers, in practice the figures make no sense.
 
paulears said:
The sensitivity between mics is often talked about, but unless the difference is severe (as in some dynamics) in practice the extra gnats whisker on the gain control makes no difference at all. I often wonder if the ONLY time we look at the specs is when we perceive a problem. You stick it in, you turn it up and if the result is a success, we are happy.
Click to expand...

What Paul said x10

Honestly, I'd recommend you buy NEITHER of those shotgun mics.

Save a bundle and get yourself a Deity S Mic 2 instead, then you can put the saved pennies towards buying a properly premium professional shotgun rather than a run of the mill medium grade mic like the NTG3/416 (that for instance is a common pairing I do, the CS3e is the one I reach for first, but on a rainy day or any other risky situations I'll put the Deity S Mic 2 in harms way instead). Or you could put that money into a hypercardioid, like one from Audix or AKG. That would give you a better pair of mics than only owning an old 416!
 
Last edited:
Samuel Dilworth said:
Thinking about this some more, it seems to me that the recorder preamp is unlikely to add significant noise with any of these mics although a dynamic mic would obviously be a different story.

The self-noise versus sensitivity figures remain intriguing. Starting to think the self-noise must somehow account for sensitivity differences, based on looking at specs of various mics I’m vaguely familiar with.
Click to expand...

Yes, self-noise measurement factors in sensitivity already so that normally (i.e. with a condenser mic and a good preamp) you don't need to consider this additionally.

Cheers,

Roland
 
You must log in or register to reply here.
Back
Top