AnalogSteph
Major Contributor
...as long as it wasn't truly and utterly terrible, of course. Well, so much for that.
I stumbled across this quite by accident while struggling with the perils of inexpensive audio interfaces:
www.audiosciencereview.com
(TL;DR: This UA-2x2 specimen has a bad left mic input channel that goes noisy when phantom power is on (probably a bad coupling cap), and remnants of USB power supply voltage noise on the PC I needed this for still make it into phantom power on the UR12, fixed by using an external charger. Quite the ordeal.)
In any case, I found that low-noise condenser mics did, in fact, exhibit audibly more white noise on these interfaces with their rather mediocre EIN - I estimate it to be about 118 dBu flat @ 24 kHz BW on the UR12, so perhaps about -121 dBu(A), which I would have considered mediocre but irrelevant. Not necessarily night and day but it's audible in a direct comparison with e.g. my Mackie 402 VLZ4 mixer, with the mic's own noise floor being almost entirely 1/f (Brown noise, about 20 dB / decade) and much more gentle on the ears. It's also quite obvious in an FFT, one will bottom out starting at a few kHz and the other just keeps on descending for a good while longer. 6 dB, easily.
Good candidates to demonstrate this are not necessarily expensive - a t.bone SC 400 from Thomann's house brand does the job (32 mm LDC, self-noise spec 18 dB(A), sens about -33 dBV/Pa), as does a Marantz Pro MPM-1000 in the same price class (18 mm "LDC", more like MDC, self-noise spec 17 dB(A), sens -38 dBV/Pa and less proximity effect, so a fair bit more critical still). As such, the same is likely to be the case for the usual low-noise LDC suspects, as well as the odd SDC if it's very good in that respect (maybe a Rode NT5 or the like).
I do have some mics that will utterly swamp even a mediocre input, like an old Samson CL8 multipattern LDC that I suspect generates 20-23 dB(A) worth of self-noise, with a fair bit that's approaching white (that's inexpensive condensers from 15-20 years ago for ya). Likewise, the t.bone SC 140 with fundamentally better-behaved electronics but a smaller capsule doesn't quite cut the mustard either, there's just too much 1/f noise altogether... but I mean, it's an inexpensive SDC (99€ the pair), what do you expect.
Some spectra taken with an MPM1000 wrapped in a thick blacket:
Steinberg UR12 Black, full gain, with 440 Hz caltone played on speaker,+30 dB +15.3 dB (little input level slider oopsie):
Makes it to about -81.5 dB with a following wind by 20 kHz. More like -81.
Mackie 402VLZ4 (gain 3 o'clock or a tad past +40 dB, tone flat, ch1 level 2 o'clock or ~+3 dB, main mix about -12 dB) into onboard Realtek ALC1200 on Asus TUF Z590-PLUS WIFI, with 440 Hz caltone played on speaker, +30.8 dB to match level:
Gets to about -87.0 dB by 20 kHz (note that gain isn't even maxed). Either is far away from the respective ADC limits. This here would be the white noise equivalent of -84.8 dBFS, and the input makes it to -105/106 dBFS equivalent up there with channel level at minimum.
(EDIT: Input level was also turned down here, -7.0 dB in this case. This is getting silly. Anyway, levels were matched, so the comparison is still valid.)
Peculiarly, there is a quite a bit more rumble going on when using the UR12, this time on USB using my home PC. Let me just grab a USB charger quick...
Nope, doesn't make a huge difference here. You can make out some remains of the rumble on internal power, so in that sense external power is a tad quieter, but it's not that much on this particular machine, unlike my office PC.
Ext Power:
Bus Power:
You can download the files to listen to here.
For fun, here's what the SC 400 looks like on the 402 VLZ4 with comparable settings (no blanket though):
Despite being a fair bit hotter, it's actually 4-6 dB quieter throughout the midrange and not any worse in the treble. (In sum, roughly 4 dB A-weighted less noise. And it's about 5 dB hotter, plus more proximity effect.) Kind of what you expect in a "real" 1" condenser vs. 18 mm electret capsule, mind you.
I stumbled across this quite by accident while struggling with the perils of inexpensive audio interfaces:
Poking around inside a Swissonic UA-2x2
Can a 2x2, 24/96 capable (UAC1) audio interface for a mere 65€ from Thomann's house brand be any good? Well, I just happened to need something to take care of an XLR condenser mic so I could finally liberate the Audient EVO 4 from my office PC and use it for portable measurements again, and I...
www.audiosciencereview.com
In any case, I found that low-noise condenser mics did, in fact, exhibit audibly more white noise on these interfaces with their rather mediocre EIN - I estimate it to be about 118 dBu flat @ 24 kHz BW on the UR12, so perhaps about -121 dBu(A), which I would have considered mediocre but irrelevant. Not necessarily night and day but it's audible in a direct comparison with e.g. my Mackie 402 VLZ4 mixer, with the mic's own noise floor being almost entirely 1/f (Brown noise, about 20 dB / decade) and much more gentle on the ears. It's also quite obvious in an FFT, one will bottom out starting at a few kHz and the other just keeps on descending for a good while longer. 6 dB, easily.
Good candidates to demonstrate this are not necessarily expensive - a t.bone SC 400 from Thomann's house brand does the job (32 mm LDC, self-noise spec 18 dB(A), sens about -33 dBV/Pa), as does a Marantz Pro MPM-1000 in the same price class (18 mm "LDC", more like MDC, self-noise spec 17 dB(A), sens -38 dBV/Pa and less proximity effect, so a fair bit more critical still). As such, the same is likely to be the case for the usual low-noise LDC suspects, as well as the odd SDC if it's very good in that respect (maybe a Rode NT5 or the like).
I do have some mics that will utterly swamp even a mediocre input, like an old Samson CL8 multipattern LDC that I suspect generates 20-23 dB(A) worth of self-noise, with a fair bit that's approaching white (that's inexpensive condensers from 15-20 years ago for ya). Likewise, the t.bone SC 140 with fundamentally better-behaved electronics but a smaller capsule doesn't quite cut the mustard either, there's just too much 1/f noise altogether... but I mean, it's an inexpensive SDC (99€ the pair), what do you expect.
Some spectra taken with an MPM1000 wrapped in a thick blacket:
Steinberg UR12 Black, full gain, with 440 Hz caltone played on speaker,
Makes it to about -81.5 dB with a following wind by 20 kHz. More like -81.
Mackie 402VLZ4 (gain 3 o'clock or a tad past +40 dB, tone flat, ch1 level 2 o'clock or ~+3 dB, main mix about -12 dB) into onboard Realtek ALC1200 on Asus TUF Z590-PLUS WIFI, with 440 Hz caltone played on speaker, +30.8 dB to match level:
Gets to about -87.0 dB by 20 kHz (note that gain isn't even maxed). Either is far away from the respective ADC limits. This here would be the white noise equivalent of -84.8 dBFS, and the input makes it to -105/106 dBFS equivalent up there with channel level at minimum.
(EDIT: Input level was also turned down here, -7.0 dB in this case. This is getting silly. Anyway, levels were matched, so the comparison is still valid.)
Peculiarly, there is a quite a bit more rumble going on when using the UR12, this time on USB using my home PC. Let me just grab a USB charger quick...
Nope, doesn't make a huge difference here. You can make out some remains of the rumble on internal power, so in that sense external power is a tad quieter, but it's not that much on this particular machine, unlike my office PC.
Ext Power:
Bus Power:
You can download the files to listen to here.
For fun, here's what the SC 400 looks like on the 402 VLZ4 with comparable settings (no blanket though):
Despite being a fair bit hotter, it's actually 4-6 dB quieter throughout the midrange and not any worse in the treble. (In sum, roughly 4 dB A-weighted less noise. And it's about 5 dB hotter, plus more proximity effect.) Kind of what you expect in a "real" 1" condenser vs. 18 mm electret capsule, mind you.
Last edited:
