Topic: Are preamps for pu$$ies?

[jerryfreak]

just wondering about your thoughts on this.

My new rig doesnt have a pre, and it sounds great to my ears (MK4v>CMR>battery box>AD2K)

normalizing once in soundforge at about about +25 - +30 dB doesnt seem to hurt the sound at all.

While minimalist, this is actually my new 'open' rig. I'd love love love to find a super small, super low noise floor stealth AD to eschew the gain stage entirely while stealthing. Whose got time to check levels when you're watching your back? the only good small AD of fthe top of my head is the sbm-1, but its only 16bit, and wouldnt have the headroom to pull this off.

i was hoping to get some feedback on this setup from the more electrically inclined. as to any major disadvantages with regard to headroom, dynamics, etc.

it seems with an AD capable of a clean -117db, and a mic with a noise floor of -75 to -80, id never be limited.

check it out. AD2K was set to its highest input level (+14 dB, minimum attenuation):

http://www.sendspace.com/file/uzt783

[Scooter]

so you're saying that you need to raise the level by 20-30db after the fact to get it up to a listenable level??  What are you're meters reading on the ad2k while recording??

Edit: the samples do not sound overly noisy, but man, that is a really low signal to feed the recorder.  and could get noisy as stated below.

[fivefishdiy]

Lots of Features, Small Size, Cheap Price .... you can have any (2).  But not all (3) at the same time. 

normalizing once in soundforge at about about +25 - +30 dB doesnt seem to hurt the sound at all.
But normalizing not only increases the level of your sound, it also increases the noise by the same amount. But if your music material is loud enough to mask the noise, then you'd probably not notice it... except for pauses or soft passages.... then the noise floor will rear its ugly head.

And just to be specific, there are "line preamplifiers" and "mic preamplifiers". Line preamp work with higher line level signals.  Mic preamps work with input signals in the millivolt range. Both are referred to as preamps... and even in the mic peamp category, you can have a simple electret mic preamp, or a standard mic preamp that can work with regular dynamic, condenser (with the appropriate 48V phantom power) or ribbon mics. 

Not all "preamps" are the same.

[jerryfreak]

to answer the poster above, the original wav peaks between -30 and -25 dB

while normalizing does increase the noise obviously, you run into the noise of the capsules well before the noise of the A/D, so regardless of where the levels are at, it all sounds the same. with this setup, once normalized you can clearly hear whispers across the room that peaked at like -70 dB (note the recording above is a noisy bar room, near a fan too)

preamp 'flavors' aside, there is always a compromise when you amplify anything (distortion). IMO, the noise introduced by doing it digitally is the same (or less) than a separate physical gain stage.

What i am concerend about, is the digital theory of defining a waveform at levels that low.

in other words, are their less bits available to define the waveform when recording in the range of -30 to -110 dB as opposed to 0 to -80 dB? or is it essentially all the same as youre above the analog noise fllor of the preamp, which would act as a natural dither? I'm thinking about how poor 8 bit audio sounds when recorded low and then normalized. obviously the bit depth of 24bit is all about increased resolution at lower levels, i just down want to lose any subtle transients or dynamics of soft passages.

in tests, it sounds the same to my ear as using a pre (preamp flavors aside)

Lots of Features, Small Size, Cheap Price .... you can have any (2).  But not all (3) at the same time. 

normalizing once in soundforge at about about +25 - +30 dB doesnt seem to hurt the sound at all.
But normalizing not only increases the level of your sound, it also increases the noise by the same amount. But if your music material is loud enough to mask the noise, then you'd probably not notice it... except for pauses or soft passages.... then the noise floor will rear its ugly head.

And just to be specific, there are "line preamplifiers" and "mic preamplifiers". Line preamp work with higher line level signals.  Mic preamps work with input signals in the millivolt range. Both are referred to as preamps... and even in the mic peamp category, you can have a simple electret mic preamp, or a standard mic preamp that can work with regular dynamic, condenser (with the appropriate 48V phantom power) or ribbon mics. 

Not all "preamps" are the same.

[jerryfreak]

Lots of Features, Small Size, Cheap Price .... you can have any (2).  But not all (3) at the same time. 

sweet, i dont need features! Brew me up a small, cheap HQ AD, please!

FYI, here are the specs on the gear:

mics:

Sensitivity:
ca. 0.6 times the value as measured with a standard CMC microphone amplifier. For example with the MK 4 cardioid, the sensitivity is about 8 mV/Pa, the maximum sound pressure is 130 dB SPL and the equivalent input noise is
29 dB CCIR** or 18 dB A-weighted*.

Maximum output voltage:
900 mV with 20 kOhm load impedance (ca. -1 dBV)
560 mV with 2 kOhm load impedance (ca. -5 dBV)

Output impedance:    15 Ohms at 1 kHz

A/D:
THD + N = -107 dBFS (0.00033%) at -1 dBFS
Dynamic range = 117 dB, A weighted
Input level for 0dBFS = +14 to +24 dBu - switch selectable
Input Impedance = not specified.

By my logic, with a close to 900mV output from the mic, the mic would reach its max spl of 134 dB somewhere between FSD and -10 dB - in other words, you would never have to set a level with this setup, and it offers the full dynamic range of the mic. To reiterate, yes there is a noise floor there, but it is the mic capsule itself. so for a given sound source volume/mic location, the noise on your recording will be the same regardless of if you are recording at 0db or -30 dB and post-normalizing.

[ethan]

Normalizing 24bit from -30 is probably fine but you  need to have your ears checked if you don't hear a difference on 16bit sources.

[jerryfreak]

yes, this wouldnt work with a 16-bit AD, and i would not attempt that

Normalizing 24bit from -30 is probably fine but you  need to have your ears checked if you don't hear a difference on 16bit sources.

[Nick's Picks]

I dont know...., add a preamp and then see if its "no difference".

[fivefishdiy]

By my logic, with a close to 900mV output from the mic, the mic would reach its max spl of 134 dB somewhere between FSD and -10 dB - in other words, you would never have to set a level with this setup, and it offers the full dynamic range of the mic. To reiterate, yes there is a noise floor there, but it is the mic capsule itself. so for a given sound source volume/mic location, the noise on your recording will be the same regardless of if you are recording at 0db or -30 dB and post-normalizing.

I don't think there are any gear with 20K input impedance... so I would use the lower value of 560mV at 2K load. That's more realistic.

By my logic, with a close to 900mV output from the mic, the mic would reach its max spl of 134 dB somewhere between FSD and -10 dB

If I understand your sentence right, you're looking at this backwards and equating apples to oranges.

The max SPL of the mic is the loudest level it can be subjected to. (And note that 130dB is 10dB more than the threshold of hearing pain, 120dB. )

134dB SPL is not the output level of the mic. The mic will never "reach" this 130dB SPL.
SPL = sound pressure level. The max SPL the mic can withstand... NOT the mic can reach.



the noise on your recording will be the same regardless of if you are recording at 0db or -30 dB and post-normalizing.

No, it's not the same.

Normalizing will multiply by a factor of X both the signal and noise. Your overall volume level may be up, but your dynamic range is still the same if you normalize it.

Amplifying the signal while keeping noise down not only increases your volume level but also your dynamic range. This is not something "normalizing" can give you. If it did, all pro recording engineers out there will be normalizing instead of working hard to reduce noise.

Earlier you said, but its only 16bit, and wouldnt have the headroom to pull this off.

I think you meant to say "dynamic range."
As 16 bit will only have 96dB dynamic range, well below the 120dB dynamic range of human hearing.

Headroom has got nothing to do with 16 or 24 bits or dynamic range. Headroom is NOT THE SAME as Dynamic range.

"Headroom" is the maximum voltage level that your gear can work with without clipping, and is measured in dBu (note the "u", it's not just dB).   Where 0dBu = 0.775Vrms = 2.45Vpp

[Church-Audio]

Lots of Features, Small Size, Cheap Price .... you can have any (2).  But not all (3) at the same time. 

sweet, i dont need features! Brew me up a small, cheap HQ AD, please!

FYI, here are the specs on the gear:

mics:

Sensitivity:
ca. 0.6 times the value as measured with a standard CMC microphone amplifier. For example with the MK 4 cardioid, the sensitivity is about 8 mV/Pa, the maximum sound pressure is 130 dB SPL and the equivalent input noise is
29 dB CCIR** or 18 dB A-weighted*.

Maximum output voltage:
900 mV with 20 kOhm load impedance (ca. -1 dBV)
560 mV with 2 kOhm load impedance (ca. -5 dBV)

Output impedance:    15 Ohms at 1 kHz

A/D:
THD + N = -107 dBFS (0.00033%) at -1 dBFS
Dynamic range = 117 dB, A weighted
Input level for 0dBFS = +14 to +24 dBu - switch selectable
Input Impedance = not specified.

By my logic, with a close to 900mV output from the mic, the mic would reach its max spl of 134 dB somewhere between FSD and -10 dB - in other words, you would never have to set a level with this setup, and it offers the full dynamic range of the mic. To reiterate, yes there is a noise floor there, but it is the mic capsule itself. so for a given sound source volume/mic location, the noise on your recording will be the same regardless of if you are recording at 0db or -30 dB and post-normalizing.

I wont try and teach you about audio  Or preamps because l I know you already know all that stuff. But one of the reason why your recordings are so good is because the self noise on the mics your using and your signal path is very low and your source was very loud thus your signal to noise was very good at the settings needed to obtain the level to capture the show.  Adding a good preamp will mean your signal to noise actually improves but in some situations in some recordings that's not something you might notice. It all depends on the level of your source. But good preamps also add flavor to the signal something I know you are aware of. I think in the end there are times when you dont need one and there are times when you do. But the nature of the concert taper is one of never knowing for sure who is mixing the live sound and how loud it will be to a certainty. Thus having a preamp available at all times is a good thing provided its not taking away all the sweetness of your great mics.

Chris

[stevetoney]

Thus having a preamp available at all times is a good thing provided its not taking away all the sweetness of your great mics.

So can I therefore conclude, Chris, that since I use preamps, I'm not a pussy?   

Bumper sticker idea...Preamps are for REAL men!  [Notice font color is pink.]   

[Church-Audio]

Thus having a preamp available at all times is a good thing provided its not taking away all the sweetness of your great mics.

So can I therefore conclude, Chris, that since I use preamps, I'm not a pussy?   

Bumper sticker idea...Preamps are for REAL men!  [Notice font color is pink.]   

Thats right man... Preamps are for real men.. Who like it "hot"

Thank you I will be here all week folks... Thanks for comming out! 

[jerryfreak]

I don't think there are any gear with 20K input impedance... so I would use the lower value of 560mV at 2K load. That's more realistic.

thanks, that actually jives better with my observations

By my logic, with a close to 900mV output from the mic, the mic would reach its max spl of 134 dB somewhere between FSD and -10 dB

If I understand your sentence right, you're looking at this backwards and equating apples to oranges.

The max SPL of the mic is the loudest level it can be subjected to. (And note that 130dB is 10dB more than the threshold of hearing pain, 120dB. )

134dB SPL is not the output level of the mic. The mic will never "reach" this 130dB SPL.
SPL = sound pressure level. The max SPL the mic can withstand... NOT the mic can reach.

yes i understand that. My point was that the (lack of) gain strategy allows for the full range of the mic, and that the mic would suffer input distortion before the rest of the signal chain.

the noise on your recording will be the same regardless of if you are recording at 0db or -30 dB and post-normalizing.

No, it's not the same.

Normalizing will multiply by a factor of X both the signal and noise. Your overall volume level may be up, but your dynamic range is still the same if you normalize it.

Amplifying the signal while keeping noise down not only increases your volume level but also your dynamic range. This is not something "normalizing" can give you. If it did, all pro recording engineers out there will be normalizing instead of working hard to reduce noise.

i have to disagree with you on this one  For this example, im going to refer to the self-noise spec of the mic as opposed to dynamic range, please correct me if im wrong in any assumptions here.) Ok here goes. The signal coming off the mic is what it is. The noise coming off the mic is what it is. These are in a fixed ratio (maybe not ratio, but fixed dBA difference lets say) This is why placement is so important, to get the highest level possible into your mic and keep the noise down. The noise level of the mic relative to the input signal is entirely independent of ANYTHING you put downstream. Lets assume my source has a maximum volume level of 100 dBA (typical not-too-loud concert environment) and the mic noise is 18 dBA . If i use a preamp and record with peaks at 0dB, the self noise of the mic is at -80 dB. If i dont use a preamp, and my peaks are at -30, the self noise of the mics is recorded at -110 dB. When normalizing, you amplify the following three signal (or noise) levels):

1. peaks get normalized from -30 to 0 dB
2. self noise of mic gets amplified from -110 to -80 dB (the same as if you had used a preamp for gain)
3. self noise of the A/D gets amplified from -117 dBA to -87 dBA (inaudible? relative to mic noise)
4. the 'quantization noise'? (digital noise floor, LSB, whatever) gets amplified from -144 dB to -114 dB and subbsequently dithered - absolutely inaudible

With this setup, no matter what you do, i beleive it is impossible to create additional audible noise that wouldnt be masked from the mics in the first place. in other words, regardless of what gain you use, if your having noise problems, its because the volume of your source wasnt high enough and the self noise of your mic is controlling. i have tried to hear the noise floor of he ad2k, i cannot distinguish it from the noise floor of the mics themselves.

Lets take an extreme case: peaking at -70 dB vs using 70 dB of gain in the field - all youre gonna hear is predominantly the noise of the mic, and thats a mic placement issue, entirely independent of signal chain - 'garbage in, garbage out.

Earlier you said, but its only 16bit, and wouldn't have the headroom to pull this off.

I think you meant to say "dynamic range."
As 16 bit will only have 96dB dynamic range, well below the 120dB dynamic range of human hearing.

Headroom has got nothing to do with 16 or 24 bits or dynamic range. Headroom is NOT THE SAME as Dynamic range.

"Headroom" is the maximum voltage level that your gear can work with without clipping, and is measured in dBu (note the "u", it's not just dB).   Where 0dBu = 0.775Vrms = 2.45Vpp

thanks, yes i was speaking of dynamic range. i kno wim not the first one to brutally use those words interchangably

[jerryfreak]

I wont try and teach you about audio  Or preamps because l I know you already know all that stuff. But one of the reason why your recordings are so good is because the self noise on the mics your using and your signal path is very low and your source was very loud thus your signal to noise was very good at the settings needed to obtain the level to capture the show.  Adding a good preamp will mean your signal to noise actually improves

again, this is where i disagree, sure a preamp would boost the signal, but it would also boost the self noise of the mic equally - and thats the limiting factor in this signal chain.

but in some situations in some recordings that's not something you might notice. It all depends on the level of your source. But good preamps also add flavor to the signal something I know you are aware of. I think in the end there are times when you dont need one and there are times when you do. But the nature of the concert taper is one of never knowing for sure who is mixing the live sound and how loud it will be to a certainty. Thus having a preamp available at all times is a good thing provided its not taking away all the sweetness of your great mics.

again, i think the extremely low noise floor of the AD2K offers me the full dynamic range of the mics , even at quiet levels, so unless i want a 'flavor', i theoretically would never need a preamp.

as i stated before, no gain is perfect. if you us a pre, there will be distortion (measurable, not necessarily audible). by doing a single normalization, there will be distortion in the form of rounding errors (also measurable, not necessarily audible). i propose that even considering the rounding errors, a simple digital algorithm of normalization would introduce less distortion than ANY analog audio component.

[jerryfreak]

LOL - my obviously inflammatory post title is attracting attention and generating the traffic on this thread that i had hoped for.

I still think ALL you guys are too cool for skool!

Thus having a preamp available at all times is a good thing provided its not taking away all the sweetness of your great mics.

So can I therefore conclude, Chris, that since I use preamps, I'm not a pussy?   

Bumper sticker idea...Preamps are for REAL men!  [Notice font color is pink.]