|
Post by mike on Oct 3, 2020 14:47:16 GMT -6
Not sure why the drop box links said the files were deleted when clicking on them, but guessing that it's something to do with my end ,..... but I could listen to the soundcloud links. Thanks very much for sharing, much appreciated.
|
|
|
Post by EmRR on Oct 3, 2020 15:25:24 GMT -6
Not sure why the drop box links said the files were deleted when clicking on them, but guessing that it's something to do with my end ,..... but I could listen to the soundcloud links. Thanks very much for sharing, much appreciated. Because I linked back to a test from 4 years ago, in another thread.
|
|
|
Post by Ward on Oct 3, 2020 20:09:16 GMT -6
Listening back... my reactions haven't changed. The DPA 4011 and the Soyuz were clearly the clearest and had the best midrange detail with pleasant presence. I still think a KM84 will beat all of them.
|
|
|
Post by mdmitch2 on Oct 3, 2020 23:04:32 GMT -6
Not sure why the drop box links said the files were deleted when clicking on them, but guessing that it's something to do with my end ,..... but I could listen to the soundcloud links. Thanks very much for sharing, much appreciated. Yeah sorry I must have deleted the files from Dropbox a while back. Glad the SoundCloud link still works, even if the quality isn’t the best.
|
|
|
Post by bricejchandler on Oct 4, 2020 1:07:43 GMT -6
Listening back... my reactions haven't changed. The DPA 4011 and the Soyuz were clearly the clearest and had the best midrange detail with pleasant presence. I still think a KM84 will beat all of them. Agreed! I use Schoeps MK4s for my acoustics which I love but I picked the Soyuz and the DPA blind ( though I did also pick out the MK4s which I have a pretty unmistakable sound). Soyuz always sound great in tests when I hear them, I really have to try some out. A little more modern and exciting compared to the Schoeps sound, seems like it would be a great compliment.
|
|
|
Post by Martin John Butler on Oct 4, 2020 8:21:59 GMT -6
I've used both the Soyuz 0-11 and 0-13, I prefer the 0-13 (FET). I've used the KM84. To me, the Soyuz is what Neumann would do if they had updated the 84. The 0-13 has way more gain, and it's super clean. Very similar tone, but a bit more presence on the Soyuz. Perhaps the joy of the KM84 is that all the specs add up to a track that easily sits in a mix, and has the best off-axis tone of all. I would say it's a very tough call if you handed me both and said pick one, and it's yours.
|
|
|
Post by Ward on Oct 4, 2020 9:06:40 GMT -6
I've used both the Soyuz 0-11 and 0-13, I prefer the 0-13 (FET). I've used the KM84. To me, the Soyuz is what Neumann would do if they had updated the 84. IThe 0-`13 has way more gain, and it's super clean. Very similar tone, but a bit more presence on the Soyuz. Perhaps the joy of the KM84 is that all the specs add up to a track that easily sits in a mix, and has the best off-axis tone of all. I would say it's a very tough call if you handed me both and said pick one, and it's yours. but what's the cleanest way to get more gain for quiet sources? 1. Jack the mic? 2. Add a line booster? 3. 80db of preamp gain? Which method would produce the least amount of noise(s)? Asking our resident builders . . . jsteiger? dandeurloo? EmRR? svart? audioscape? matt@IAA? bluegrassdan? @dckelly? so many more I could ask . . .
|
|
|
Post by Martin John Butler on Oct 4, 2020 9:34:20 GMT -6
The Cloud Lifter worked beautifully on my friend's mic. I'm not sure, but I think it was the Cascade Vin-Jet with Lundahl transformer.
|
|
|
Post by matt@IAA on Oct 4, 2020 9:45:12 GMT -6
To get the lowest noise, your chain’s first amplifier should always have the lowest noise figure and highest gain. Going solely by that it’s better to use a quiet preamp with a second amp stage behind it. The only caveats I can think of are long cables or impedance mismatch effects.
|
|
|
Post by svart on Oct 4, 2020 10:07:39 GMT -6
I've used both the Soyuz 0-11 and 0-13, I prefer the 0-13 (FET). I've used the KM84. To me, the Soyuz is what Neumann would do if they had updated the 84. IThe 0-`13 has way more gain, and it's super clean. Very similar tone, but a bit more presence on the Soyuz. Perhaps the joy of the KM84 is that all the specs add up to a track that easily sits in a mix, and has the best off-axis tone of all. I would say it's a very tough call if you handed me both and said pick one, and it's yours. but what's the cleanest way to get more gain for quiet sources? 1. Jack the mic? 2. Add a line booster? 3. 80db of preamp gain? Which method would produce the least amount of noise(s)? Asking our resident builders . . . jsteiger? dandeurloo? EmRR? svart? audioscape? matt@IAA? bluegrassdan? @dckelly? so many more I could ask . . . The least noise is always more gain further towards the source, so a hotter mic output is best.
|
|
|
Post by EmRR on Oct 4, 2020 11:12:55 GMT -6
but what's the cleanest way to get more gain for quiet sources? 1. Jack the mic? 2. Add a line booster? 3. 80db of preamp gain? Which method would produce the least amount of noise(s)? To get the lowest noise, your chain’s first amplifier should always have the lowest noise figure and highest gain. Going solely by that it’s better to use a quiet preamp with a second amp stage behind it. ....and no inline booster has a lower noise figure than the average preamp, maybe some ancients, but not most. Then, the two noise figures are additive. Jim Williams put some numbers to it at GS in a thread somewhere, looking at the Cloudlifter with an Audio Precision; it was worse than without. It's counterintuitive, but as I keep reading various places (and there it is again in the quote above), the lowest noise figure with typical gain controlled by negative feedback is max gain. The noise SEEMS louder because of the GAIN, but if you subtract the GAIN it tends to be lowest. If noise shows, generally you're up against realistic limits of source/mic more than anything else. In most cases acoustic residual noise (air motion, etc) should swamp electronic noise with an open mic anyway. Hi-gain input transformers help, lots of my 1940's tube gear seems quieter with a ribbon than some modern transformerless thing, or a thing like a Neve with very low transformer gain. An API/Q8 is better in that regard, but then problematic with loud sources, the hatred of the input pad, etc. LDC's higher output helps, if it's appropriate. Sennheiser MKH series mics have a leg up on most others too, being RF SDC's with higher capsule output levels. If I can't get a quiet pickup with an MKH mic, it's an impossible request. People make recordings of crickets and birdsong in quiet windless fields with those mics.
|
|
|
Post by svart on Oct 4, 2020 14:45:44 GMT -6
but what's the cleanest way to get more gain for quiet sources? 1. Jack the mic? 2. Add a line booster? 3. 80db of preamp gain? Which method would produce the least amount of noise(s)? To get the lowest noise, your chain’s first amplifier should always have the lowest noise figure and highest gain. Going solely by that it’s better to use a quiet preamp with a second amp stage behind it. ....and no inline booster has a lower noise figure than the average preamp, maybe some ancients, but not most. Then, the two noise figures are additive. Jim Williams put some numbers to it at GS in a thread somewhere, looking at the Cloudlifter with an Audio Precision; it was worse than without. It's counterintuitive, but as I keep reading various places (and there it is again in the quote above), the lowest noise figure with typical gain controlled by negative feedback is max gain. The noise SEEMS louder because of the GAIN, but if you subtract the GAIN it tends to be lowest. If noise shows, generally you're up against realistic limits of source/mic more than anything else. In most cases acoustic residual noise (air motion, etc) should swamp electronic noise with an open mic anyway. Hi-gain input transformers help, lots of my 1940's tube gear seems quieter with a ribbon than some modern transformerless thing, or a thing like a Neve with very low transformer gain. An API/Q8 is better in that regard, but then problematic with loud sources, the hatred of the input pad, etc. LDC's higher output helps, if it's appropriate. Sennheiser MKH series mics have a leg up on most others too, being RF SDC's with higher capsule output levels. If I can't get a quiet pickup with an MKH mic, it's an impossible request. People make recordings of crickets and birdsong in quiet windless fields with those mics. Re: negative feedback noise. The feedback resistors add thermal noise. The value matters a lot because of this. You want to keep the values low, but not so low that the amp starts wasting power, then it is even more noisy. The dividing resistors also contribute due to the input pin currents but to a lesser extent. Anyway, system NF is roughly equivalent to the amount of amplifier self-noise plus the resistor noise plus the cumulative loss before it. Any impedance matching, cabling/connector losses, filters, active components and circuits, etc., add their through-loss as noise.
|
|
|
Post by EmRR on Oct 4, 2020 15:16:02 GMT -6
Re: negative feedback noise. The feedback resistors add thermal noise. The value matters a lot because of this. You want to keep the values low, but not so low that the amp starts wasting power, then it is even more noisy. The dividing resistors also contribute due to the input pin currents but to a lesser extent. Anyway, system NF is roughly equivalent to the amount of amplifier self-noise plus the resistor noise plus the cumulative loss before it. Any impedance matching, cabling/connector losses, filters, active components and circuits, etc., add their through-loss as noise. If you look at old API data, you see earlier on they used higher value feedback resistors, they cut it in half at some point. In a lot of late '60's-early '70's gear it's normal to find more expensive IRC precision resistors in those positions, with all else being standard carbon. If something ancient is using carbon there, that's an instant noise upgrade, easily.
|
|
ericn
Temp
Balance Engineer
Posts: 14,976
|
Post by ericn on Oct 4, 2020 15:36:35 GMT -6
Since this thread is about SDC’s not ribbons and boosters and noise has come up, I’m going to point out the obvious, to obvious for my far more technically inclined friends haven’t pointed it out. With a condenser or active ribbon the pre amp inside the mic is the first gain stage. So everything after that is going to multiply it’s noise! It’s hard to think this way but even a transformer coupled ribbon or dynamic is probably getting some gain and noise from the transformer. So your pre or booster is really a second gain stage not the first.
|
|
|
Post by svart on Oct 4, 2020 15:42:32 GMT -6
Re: negative feedback noise. The feedback resistors add thermal noise. The value matters a lot because of this. You want to keep the values low, but not so low that the amp starts wasting power, then it is even more noisy. The dividing resistors also contribute due to the input pin currents but to a lesser extent. Anyway, system NF is roughly equivalent to the amount of amplifier self-noise plus the resistor noise plus the cumulative loss before it. Any impedance matching, cabling/connector losses, filters, active components and circuits, etc., add their through-loss as noise. If you look at old API data, you see earlier on they used higher value feedback resistors, they cut it in half at some point. In a lot of late '60's-early '70's gear it's normal to find more expensive IRC precision resistors in those positions, with all else being standard carbon. If something ancient is using carbon there, that's an instant noise upgrade, easily. True. Carbon-composition have crystal structures in the carbon that are horribly noisy, but also add small distortions when large voltages floor through them, making them a favorite for guitar amps, but also given an unrealistic amount of credit in doing so.. Carbon-film resistors (the modern standard light tan ones) are magnitudes better, and as we know, metal film overall is the best of the generic styles, with the thin-film variation of them being the best of all cheap resistors. Some thin film resistors easily beat wirewound resistors in self noise (and price, but not power handling).
|
|
|
Post by bluegrassdan on Oct 5, 2020 9:57:20 GMT -6
Physics determines that the lowest equivalent input noise (EIN) is around -131 dB for about 150 ohms resistance. An SSL pre, for example, may measure just a few dB higher than this, which is a very good spec. Even a well designed tube preamp can do better than -124 dB. Anything better than -120dB is fine.
|
|
|
Post by bluegrassdan on Oct 5, 2020 10:05:30 GMT -6
And, as someone mentioned earlier, EIN is usually determined and stated at a particular value, and often at or close to max gain.
Once you start attenuating is when preamp topologies will start showing greater EIN discrepancies. I have measured for noise my tube preamps against a Graham Cohen design, and the GC is just way quieter at lower gains (still well below hearing threshold, but very telling). Fixed gain topologies have more potential for noise when attenuated mid-stage.
I’m sure Jeff and others know way more about this topic than me.
|
|
|
Post by Vincent R. on May 6, 2022 5:02:22 GMT -6
Forgot about this shootout. It was nice to listen to it again.
|
|
rpc
Junior Member
Posts: 67
|
Post by rpc on May 6, 2022 7:25:30 GMT -6
but what's the cleanest way to get more gain for quiet sources? 1. Jack the mic? 2. Add a line booster? 3. 80db of preamp gain? Which method would produce the least amount of noise(s)? Asking our resident builders . . . jsteiger ? dandeurloo ? EmRR ? svart ? audioscape ? matt@IAA ? bluegrassdan ? @dckelly? so many more I could ask . . . Obviously very late to this game, but my answer would be: Linear Systems LSK189. You can make arguments about the sound qualities of 2N3819/LS170 etc. but subbing in an LSK189 (and rebiasing) will drop noise 6-10dB. This assumes you want to rush in where angels fear to tread, however!
|
|
|
Post by Omicron9 on May 6, 2022 8:10:04 GMT -6
I've used both the Soyuz 0-11 and 0-13, I prefer the 0-13 (FET). I've used the KM84. To me, the Soyuz is what Neumann would do if they had updated the 84. IThe 0-`13 has way more gain, and it's super clean. Very similar tone, but a bit more presence on the Soyuz. Perhaps the joy of the KM84 is that all the specs add up to a track that easily sits in a mix, and has the best off-axis tone of all. I would say it's a very tough call if you handed me both and said pick one, and it's yours. but what's the cleanest way to get more gain for quiet sources? 1. Jack the mic? 2. Add a line booster? 3. 80db of preamp gain? Which method would produce the least amount of noise(s)? Asking our resident builders . . . jsteiger ? dandeurloo ? EmRR ? svart ? audioscape ? matt@IAA ? bluegrassdan ? @dckelly? so many more I could ask . . . One thing I've done to get more gain for quiet sources is to move the mics closer to the source. Really interesting shootout, BTW. -09
|
|
|
Post by Ward on May 6, 2022 10:04:53 GMT -6
One thing I've done to get more gain for quiet sources is to move the mics closer to the source. *all the usual snips* -09 Can you speak up please? LOL I got you.
|
|
|
Post by tahoebrian5 on May 6, 2022 10:35:05 GMT -6
This is a nice comparison. The KM184 I love for anything strummed but the Soyuz kills on the arpeggiated section.
|
|
|
Post by Martin John Butler on May 6, 2022 12:19:08 GMT -6
It is a nicely done comparison, but something's not right in this shootout. The 0-13 FET sounds brighter than it does here. The KM184 seems normal here and the Soyuz sounds much darker. The 0-13 is not a dark sounding mic at all.
|
|