Yes, what krivium said. Not fast enough for "protection". Very useful compressor, though.
We seem to have drifted a bit, and there isn't a clear understanding here of the difference between a peak limiter a clipper, a soft clipper and a compressor. None of them are the same, all different, with different results and purposes.
The primary difference is in how the level is detected and how the unit responds.
A compressor usually responds to the RMS value of a waveform, or an average value, sometimes an integrated peak value, but has a compression ratio (input change to output change) of less than 10:1. It will not control peaks as it doesn't respond to them.
A limiter determines level by sensing RMS value, or peak value, typically response faster (faster attack time), and as a ratio above 10:1, often 100:1 or even infinite. If it is RMS responding, it will not control peaks. If it is peak responding, but doesn't attack quickly, it won't control peaks.
A Peak Limiter senses the peak value only, and responds extremely quickly, sometimes measured in microseconds. It will control peaks, and has an infinite ratio or ratio higher than 100:1.
All devices can introduce distortion based on te release time. Faster release means higher distortion. This is why aggressive processing is often done with multi-band limiting or compression, as individual bands can have different attack and release times, reducing distortion.
A clipper simply clamps the maximum voltage to a fixed quantity, not to exceed. Doing so creates a hard limit, but also produces the maximum distortion above that limit.
A soft clipper is just a clipper that has a gradual threshold. It still produces high distortion, but somewhat less than a classic clipper.
All are useful, all have different applications. If the application is overload protection, or protection of speakers, a peak limiter is recommended, put more importantly, proper system design that matches speakers, amplifier and application so no elements are stressed in normal use.
In this thread, the application is blurry. It seems to be a personal desire for information, without a specific need for protection of any device.
There is also a bit of misunderstanding of metering. A VU meter is a very well defined standard meter. Everything about the meter, even the face color, is part of that standard. Technicaly, a string of LEDs with VU meter response does not meet the standard.
A VU meter indicates the average value of the input wave. It does NOT indicate the RMS value or peak value. It also does not correctly indicate the property of psychoacoustic loudness. It was the best that could be done in 1938, but was primarily intended as a standardized "volume indicator" to be used among radio networks and telephone companies. It as application in recording devices as well. VU meters are not loudness meters, as they do not include psychoacoustic loudness algorithms.
A Peak meter indicate the peak value of the input wave. It indicates this quickly, and has a slow fall-back so the indication is visible. Variations are True Peak (the most accurate and fastest), and PPM, Peak Program Meter, which is a slowed response based on the audibility of distortion over time. Peak meters do not indicate loudness.
More information on the application would be helpful.
We seem to have drifted a bit, and there isn't a clear understanding here of the difference between a peak limiter a clipper, a soft clipper and a compressor. None of them are the same, all different, with different results and purposes.
The primary difference is in how the level is detected and how the unit responds.
A compressor usually responds to the RMS value of a waveform, or an average value, sometimes an integrated peak value, but has a compression ratio (input change to output change) of less than 10:1. It will not control peaks as it doesn't respond to them.
A limiter determines level by sensing RMS value, or peak value, typically response faster (faster attack time), and as a ratio above 10:1, often 100:1 or even infinite. If it is RMS responding, it will not control peaks. If it is peak responding, but doesn't attack quickly, it won't control peaks.
A Peak Limiter senses the peak value only, and responds extremely quickly, sometimes measured in microseconds. It will control peaks, and has an infinite ratio or ratio higher than 100:1.
All devices can introduce distortion based on te release time. Faster release means higher distortion. This is why aggressive processing is often done with multi-band limiting or compression, as individual bands can have different attack and release times, reducing distortion.
A clipper simply clamps the maximum voltage to a fixed quantity, not to exceed. Doing so creates a hard limit, but also produces the maximum distortion above that limit.
A soft clipper is just a clipper that has a gradual threshold. It still produces high distortion, but somewhat less than a classic clipper.
All are useful, all have different applications. If the application is overload protection, or protection of speakers, a peak limiter is recommended, put more importantly, proper system design that matches speakers, amplifier and application so no elements are stressed in normal use.
In this thread, the application is blurry. It seems to be a personal desire for information, without a specific need for protection of any device.
There is also a bit of misunderstanding of metering. A VU meter is a very well defined standard meter. Everything about the meter, even the face color, is part of that standard. Technicaly, a string of LEDs with VU meter response does not meet the standard.
A VU meter indicates the average value of the input wave. It does NOT indicate the RMS value or peak value. It also does not correctly indicate the property of psychoacoustic loudness. It was the best that could be done in 1938, but was primarily intended as a standardized "volume indicator" to be used among radio networks and telephone companies. It as application in recording devices as well. VU meters are not loudness meters, as they do not include psychoacoustic loudness algorithms.
A Peak meter indicate the peak value of the input wave. It indicates this quickly, and has a slow fall-back so the indication is visible. Variations are True Peak (the most accurate and fastest), and PPM, Peak Program Meter, which is a slowed response based on the audibility of distortion over time. Peak meters do not indicate loudness.
More information on the application would be helpful.
Few quick things to add (not critical but hopefully background that will aid in understanding):
1. A clipper, soft or hard, limits the maximum peak value only. It does not limit RMS value. In fact, RMS value continues to rise when the input level is pushed well beyond clipping.
2. A peak limiter also does not initilly limit RMS value. Like a clipper, the RMS value will continue to rise when the input is pushed higher than the limiting threshold. It will not rise as much as when signal is pushed beyond a clipping threshold, and it will leventually stop rising as the gain control element holds the singnal back.
3. Most traditional limiters and compressors sample their output, compare to the input, and use the difference, an "error voltage", to control gain. This was mostly due to the nonlinear control characteristics of gain congrol devices like FETs, bipolar transistors, and optical devices. WIth the arrival of VCAs with well defined logarithic control laws, it became possible to design a "feed forward" limiter, where the needed gain reduction was calculated instantaneously based on the input, rather than the output. VCA based dynamics processors also can be set for fixed attack rates in dB/sec rather than a time constant where real attack and release times were level dependant.
1. A clipper, soft or hard, limits the maximum peak value only. It does not limit RMS value. In fact, RMS value continues to rise when the input level is pushed well beyond clipping.
2. A peak limiter also does not initilly limit RMS value. Like a clipper, the RMS value will continue to rise when the input is pushed higher than the limiting threshold. It will not rise as much as when signal is pushed beyond a clipping threshold, and it will leventually stop rising as the gain control element holds the singnal back.
3. Most traditional limiters and compressors sample their output, compare to the input, and use the difference, an "error voltage", to control gain. This was mostly due to the nonlinear control characteristics of gain congrol devices like FETs, bipolar transistors, and optical devices. WIth the arrival of VCAs with well defined logarithic control laws, it became possible to design a "feed forward" limiter, where the needed gain reduction was calculated instantaneously based on the input, rather than the output. VCA based dynamics processors also can be set for fixed attack rates in dB/sec rather than a time constant where real attack and release times were level dependant.
@krivium thanks the GainStage/Structure Post was VERY helpful.
I also now see I can use the SPL button on each channel of the mixer to set the gain using the VU meter. I can try to keep all my gain levels of each stage between -3db and -6db below their max 😉
I see some amps have built-in gain knobs on the inputs, so the signal can be attenuated further down the signal path ( as opposed to before the pre-amp ), reducing the signal to noise ratio. My mixer master output fader should accomplish the same.
I now need to learn about connecting my oScope to my mixer output safely. I guess I'll create a new post to get advice on what leads/probes I should use. ( I have some 1/8"->BNC probes I use to connect into eurorack synths and would need something similar for TS ( or TS->RCA ).
I also now see I can use the SPL button on each channel of the mixer to set the gain using the VU meter. I can try to keep all my gain levels of each stage between -3db and -6db below their max 😉
I see some amps have built-in gain knobs on the inputs, so the signal can be attenuated further down the signal path ( as opposed to before the pre-amp ), reducing the signal to noise ratio. My mixer master output fader should accomplish the same.
I now need to learn about connecting my oScope to my mixer output safely. I guess I'll create a new post to get advice on what leads/probes I should use. ( I have some 1/8"->BNC probes I use to connect into eurorack synths and would need something similar for TS ( or TS->RCA ).
That isn't how you adjust channel gain. You want to adjust the input gain so with the channel fader a "0" and the master at "0" you get good level on te meters without clipping the channel input. That gives you the proper control range for the channel fader and the master fader. There is never any reason to try for a particular gain setting other than to find what works. Ideally, you'd feed the highest level you can to the mixer from source devices, and use the least amount of channel input gain. But there's no gain target, just whatever works.
Functionally, yes, but not the "same thing". There is an amplifier after your master, so even if you pull the master down the final output amp is still unattenuated. Since you have a pro-level mixer and a consumer-level amplifier, the mixer's output is somewhere around 12-14dB too high for the amp. Turning te master down doesn't completely fix this even though the level might seem right, you'r paying a noise penalty. You need to pad the mixer before the power amp. The input gain control on a power amp attenuates the input signal, including whatever amplifiers are involved, just like a pad would.
Does it have to be an analog limiter, or could a digital one do? A digital limiter can work better than an analog one, since look-ahead processing is possible.
@jaddie I meant I would try to keep all the various 'stages' at -3db to -6db under each of their 'clipping point'. ( their sweet spot from what I understand ) I did not mean to imply I would adjust each channel's gain to those levels. I do understand how to adjust the channels now, so that the are at 0 when the master is at 0.
As far as the pad circuit, I think I now see what you mean. By turning the master down, I'm making the signal 'closer to the noise' and reducing the signal to noise ratio. I'll take a look at the pad circuit to attenuate -10db to -12db between the mixer and amp input.
@Markw4 yes, I am trying to retain the pure analog signal with no digital in the path...
As far as the pad circuit, I think I now see what you mean. By turning the master down, I'm making the signal 'closer to the noise' and reducing the signal to noise ratio. I'll take a look at the pad circuit to attenuate -10db to -12db between the mixer and amp input.
@Markw4 yes, I am trying to retain the pure analog signal with no digital in the path...
Yes. Digital dynamics processing solves a lot of problems with precision. Best of the best. Unless you hate those "nasty digits".
I'm fairly sure you might still not understand. For example, there's no such thing as a "sweet spot".
Excellent.
Funny. Analog isn't the pure bit. Every single analog stage has the potential for signal degradation. Digital processing with 32 bit floating point math has a dynamic range of 1528dB, which exceeds analog DR by at least 1400dB, typically more for an entire analog chain. 64 bit floating point is obviouly greater. 24 bits is unmanageable in a real world sound system, ranging from below the threshold of hearing to above the threshold of pain, and therefore, not reproducable in any studio much less living room. And with no accidential frequency response modifications, or added distortion...yeah, analog isn't the "pure" one. And there are no stairsteps, never were. I could go on, but it's like talking to a wall with this stuff. Peole hear and belive whatever they want and routinely choose to ignore science. Shocking in 2025.
@jenimitso ,
well no 18bits dacs didn't solve issues we ( professional in audio world) faced: we still had issues with dynamic range with this generation of ADC ( to have the most accuracy possible we still needed to make our gain very close to 0dbfs, which didn't solved the occasional peak/transient to clip input) and it was quickly apparent we had issue relative to math involved with digital treatments ( eq, dynamic, summing) in DAW and other digital gear.
Answer for all things treatment related was to go up in dynamic range ( number of bits involved for all treatments within software) by using things like 48bits fixed, then going floating point.
Jumping from 16bit to 24bit for converters gave opportunity to not have to be so hot on levels hitting ADC to approach 0dbfs so less ( or no) peaks was clipped during recordings ( eg i routinely have between 6 and 12 db headroom during tracking without sonic corruption, which wasn't true even with the 18bits Studer ADC/DAC we had in the studio. It doesn't stop this unit to sound sweet even by today standard but it make them less easy to use as you have to watch for levels more carefuly...).
I would not be so enthusiastic about digital as Jadie is either. Jadie is true that 'perfection' is closer to digital than most analog gear ( some are on par with the best digital gear though) but the question is: is it needed or does it fit the artistic performance?
I mean the number of people into rock/metal/variants which produce in the box and use a Ampex 2 track with gp9 'clone' plug in on masterbus (or tracks!) is incredibly high...
People swear by digital 'clone' of analog gear...
In my view it's just different kinds of tools each with pro and con. You mix and match to reach a goal. And sometimes the 'real thing' is better than any emulators ( i'm collecting old consummer reel to reel i use as hardware 'plugins' to my client surprise - no they are not Studer, 3M or Ampex- and delight, yes tape sound 'better' on some genres or instruments parts).
well no 18bits dacs didn't solve issues we ( professional in audio world) faced: we still had issues with dynamic range with this generation of ADC ( to have the most accuracy possible we still needed to make our gain very close to 0dbfs, which didn't solved the occasional peak/transient to clip input) and it was quickly apparent we had issue relative to math involved with digital treatments ( eq, dynamic, summing) in DAW and other digital gear.
Answer for all things treatment related was to go up in dynamic range ( number of bits involved for all treatments within software) by using things like 48bits fixed, then going floating point.
Jumping from 16bit to 24bit for converters gave opportunity to not have to be so hot on levels hitting ADC to approach 0dbfs so less ( or no) peaks was clipped during recordings ( eg i routinely have between 6 and 12 db headroom during tracking without sonic corruption, which wasn't true even with the 18bits Studer ADC/DAC we had in the studio. It doesn't stop this unit to sound sweet even by today standard but it make them less easy to use as you have to watch for levels more carefuly...).
I would not be so enthusiastic about digital as Jadie is either. Jadie is true that 'perfection' is closer to digital than most analog gear ( some are on par with the best digital gear though) but the question is: is it needed or does it fit the artistic performance?
I mean the number of people into rock/metal/variants which produce in the box and use a Ampex 2 track with gp9 'clone' plug in on masterbus (or tracks!) is incredibly high...
People swear by digital 'clone' of analog gear...
In my view it's just different kinds of tools each with pro and con. You mix and match to reach a goal. And sometimes the 'real thing' is better than any emulators ( i'm collecting old consummer reel to reel i use as hardware 'plugins' to my client surprise - no they are not Studer, 3M or Ampex- and delight, yes tape sound 'better' on some genres or instruments parts).
Back to roboDNA issue.
Ok so you now know how to set up gain in your desk. It'll make you closer to solve your issue.
You understood why we try to dial the master fader at UG ( Unity Gain, 0db, amplification factor of x1) while mixing.
The issue you'll face is your desk doesn't have a dedicated monitoring/Control Room section, only headphones.
On recording console we have extensive option to listen and communicate with musicians: choice of multiple monitors, dedicated level pot, choice of many sources for monitoring purpose, talkback, etc,etc,...
Yours doesn't have this option, only monitoring through headphones. It's because it's dedicated to live acts and small instalations.
So you'll have to make a choice: either you use headphone outputs to drive your amp, either the desk's main output.
Headphones out could drive the amp ( electronically speaking) but you'll have to deal with cable adaptators and the potential nasty coming with them ( crackling, noise from bad mechanical contacts, etc,etc,...).
So for the time being i would choose to use the main output and the associated fader. It's not perfect but it'll be usable and handy to learn.
So you'll need a pad. The purpose is really to scale down your console ouput ( voltage range) to your amp input ( max voltage).
But you probably don't run full blast your amp all the time... and as such probably will have to lower the master fader. It have the potential to induce nasty's but it's and acceptable compromise imho, and to be blunt i've seen myself doing things way more questionable even in multimillion dollars studios... 🙂
To help you setting up your home studio it could be nice to list your source ( synth, instruments, sound card) and your loudspeakers as well as describe physical implementation ( as Jadie pointed earlier we could help define levels -electrical and acoustic).
Ok so you now know how to set up gain in your desk. It'll make you closer to solve your issue.
You understood why we try to dial the master fader at UG ( Unity Gain, 0db, amplification factor of x1) while mixing.
The issue you'll face is your desk doesn't have a dedicated monitoring/Control Room section, only headphones.
On recording console we have extensive option to listen and communicate with musicians: choice of multiple monitors, dedicated level pot, choice of many sources for monitoring purpose, talkback, etc,etc,...
Yours doesn't have this option, only monitoring through headphones. It's because it's dedicated to live acts and small instalations.
So you'll have to make a choice: either you use headphone outputs to drive your amp, either the desk's main output.
Headphones out could drive the amp ( electronically speaking) but you'll have to deal with cable adaptators and the potential nasty coming with them ( crackling, noise from bad mechanical contacts, etc,etc,...).
So for the time being i would choose to use the main output and the associated fader. It's not perfect but it'll be usable and handy to learn.
So you'll need a pad. The purpose is really to scale down your console ouput ( voltage range) to your amp input ( max voltage).
But you probably don't run full blast your amp all the time... and as such probably will have to lower the master fader. It have the potential to induce nasty's but it's and acceptable compromise imho, and to be blunt i've seen myself doing things way more questionable even in multimillion dollars studios... 🙂
To help you setting up your home studio it could be nice to list your source ( synth, instruments, sound card) and your loudspeakers as well as describe physical implementation ( as Jadie pointed earlier we could help define levels -electrical and acoustic).
I do mix things...For a living! 😉
As i've been active during this whole evolution of digital stuff i doubt i'm mixing the technical principle at play either. 😋
Maybe you haven't read the OP issue: this is about music producing! Maybe there is better place, but robodna is here and we have the knowledge to help him! Why should we do otherwise than help him? If it bother you, then ask a mod to move it on other subforum or whatever...
Anyway i'm known to drift off topic, talk about other things that consumer gear and practice... c'est la vie! 🙂
And no 18bits dac are not the end of it all about DAC. In fact when Studer ( and other brands) released their 18bit gen of DAC it concured with the observation that most of the 'sterile cold' sound from 1st gen converters come from the simple fact that the need to run close to 0dbfs induced some nasty's both ADC side but DAC too: intersample peak happened more than often and the dacs did not handled this beautifully.
This is the time we have seen True Peak Meters appearing on the market too... they helped to deal with the issue.
And we started to try to be less demandings on dacs too ( we limited ouput to minus 0.2 or 0.1 dbfs to be more gentle with consummers dacs as a rule)...
As i've been active during this whole evolution of digital stuff i doubt i'm mixing the technical principle at play either. 😋
Maybe you haven't read the OP issue: this is about music producing! Maybe there is better place, but robodna is here and we have the knowledge to help him! Why should we do otherwise than help him? If it bother you, then ask a mod to move it on other subforum or whatever...
Anyway i'm known to drift off topic, talk about other things that consumer gear and practice... c'est la vie! 🙂
And no 18bits dac are not the end of it all about DAC. In fact when Studer ( and other brands) released their 18bit gen of DAC it concured with the observation that most of the 'sterile cold' sound from 1st gen converters come from the simple fact that the need to run close to 0dbfs induced some nasty's both ADC side but DAC too: intersample peak happened more than often and the dacs did not handled this beautifully.
This is the time we have seen True Peak Meters appearing on the market too... they helped to deal with the issue.
And we started to try to be less demandings on dacs too ( we limited ouput to minus 0.2 or 0.1 dbfs to be more gentle with consummers dacs as a rule)...
Last edited:
I'm not mixing things, but it seems you are: we are in analog line level subsection and you talk about digital... and from the start i explained there is no need for any dynamic range management in this case, explained how to cure the issue and have not moved my pov since.
It seems Robodna learned a few things along the discussion so maybe i'm not that wrong or i understood the issue he faced? Lol.
He asked to build a compressor ( which i will encourage but for other reasons and i think Ron68 and robodna should pursue the design as compressor are needed for instruments imho) but it's not needed there. You want to push him to spend money on something which is not needed?! Maybe you run a shop or something?! I don't.
The answer i propose is 4 resistors and 2 xlr... such a scandal! 🤔😱🤣
We don't know which kind of loudspeakers robodna use but i would not say N.Pass circuits are unprofessional... you obviously don't know there is a bunch of Treshold, Passlab, or even Firstwatt amplifier( i know i know how is it possible?! If i told you i setup one for the drummer headphone cues you wouldn't trust me... despite it being the truth! 🤣 ) used in pro control rooms.
If you don't believe me ask Nelson direct how many pro are part of his clients... and yes we know when something sound good usually, hence some 'hifi' brands are used in proworld... or the inverse... anyway.
It seems Robodna learned a few things along the discussion so maybe i'm not that wrong or i understood the issue he faced? Lol.
He asked to build a compressor ( which i will encourage but for other reasons and i think Ron68 and robodna should pursue the design as compressor are needed for instruments imho) but it's not needed there. You want to push him to spend money on something which is not needed?! Maybe you run a shop or something?! I don't.
The answer i propose is 4 resistors and 2 xlr... such a scandal! 🤔😱🤣
We don't know which kind of loudspeakers robodna use but i would not say N.Pass circuits are unprofessional... you obviously don't know there is a bunch of Treshold, Passlab, or even Firstwatt amplifier( i know i know how is it possible?! If i told you i setup one for the drummer headphone cues you wouldn't trust me... despite it being the truth! 🤣 ) used in pro control rooms.
If you don't believe me ask Nelson direct how many pro are part of his clients... and yes we know when something sound good usually, hence some 'hifi' brands are used in proworld... or the inverse... anyway.
Last edited:
And so what?
I never sugested anything digital, but you don't seem to want to waste your time reading the thread. I can understand though...
High power... well did he talk about driving Kinoshita's RM-7 located 4m away? Please,... like most home producer he probably listen in nearfield, so high power is probably not needed. And If he like H2 that's fine to me. I do too. Not on my monitoring though but my preference are mine, not universal in any way.
Is this preference patrol? 😉
I hope not, it doesn't make sense.
I never sugested anything digital, but you don't seem to want to waste your time reading the thread. I can understand though...
High power... well did he talk about driving Kinoshita's RM-7 located 4m away? Please,... like most home producer he probably listen in nearfield, so high power is probably not needed. And If he like H2 that's fine to me. I do too. Not on my monitoring though but my preference are mine, not universal in any way.
Is this preference patrol? 😉
I hope not, it doesn't make sense.
Last edited:
I've been down the digital/analog rabbit hole and I personally see both digital and analog being used together. There is no replacing analog for what it does, and can't be reproduced digitally. So many guitar players still don't use wireless cables to their amp for a good reason... I for one LOVE my analog synths and they have something the digital versions don't have. ( ask anyone who owns a Minimoog ) Keeping a pure analog signal for my DIY synths s is what led me to building the Nelson Pass F5m which is an absolutely amazing amp for my needs. I think humans can detect the difference between digital and analog ( they call it 'sharp/cold/clean vs smooth/warm/dirty' but can't quite put their finger on what )... I personally think it is similar to the human ear being able to detect if a tiger is moving towards you or away from you... being able to detect phasing. I also think before digital, speakers/amps were designed to distort in a way that sounded good, like guitar amps. I want both digital and analog so it's never really been a topic of debate for me. ( also, seen some analog shows like RUSH with analog equipment in the mid 80's and you don't get sound like that anymore.. very good sounding class Ds but not the same )
@krivium you are correct, I have learned QUITE A BIT about gain staging/structure.
@jadide The sweet spot I refer to is 'unity gain', so thanks to all the help, I know I am in fact correct on this 'sweet spot'... 😉
The PAD is the best solution for me from what I've learned so far, thanks. My mixer should allow me to visualize the voltage now that I understand it better.
@krivium you are correct, I have learned QUITE A BIT about gain staging/structure.
@jadide The sweet spot I refer to is 'unity gain', so thanks to all the help, I know I am in fact correct on this 'sweet spot'... 😉
The PAD is the best solution for me from what I've learned so far, thanks. My mixer should allow me to visualize the voltage now that I understand it better.
The only difference between compression and limiting is the ratio. At 20:1 or infinity:1 it is pure limiting. Clipping always causes distortion even in the digital domain. There are DSP tricks to improve it but that comes with latency. There is nothing better then having headroom and paying careful attention to the meters on the board. Small brief peaks won't break anything. It takes a dense high level signal to damage speakers. An RMS sensor in the side chain is perfectly fine to protect speakers. It sounds to me like the goal is to prevent an accidental huge signal from causing damage and the cheap DBX unit will do just that.
I'm sorry, this is incorrect. I've already discussed this in depth in post #43. Compression is made deliberately slower, and very often with different types of detectors (RMS vs peak). Yes, ratio is a key to limiting, but far from the only defining aspect.
What's not commonly realized is that the precision and solid "brick wall" effect of clipping is thrown out once the signal is band limited by reconstruction filtering. You won't see it inside the DAW, but it will be there outside. But all of this is unimportant to the OP.
I'm frankly not sure that this is about speaker protection. He has a 25 watt amp and 200 W capable speakers. He seems obsessed with the output voltage of his desk. A simple pad fixes that.
The "sweet spot" is where you have the input gain set so you get so all successive controls are close to zero (which is never unity internally). If you set everything for actual unity gain, that will not happen.
Unity gain= no amplifier gain or loss.
Fader "0" is not unity gain, because a fader always attenuates, its a lossy device. Unity gain for a fader is at the top of its travel. If that's shown as +12dB, then there must be 12dB of gain after the fader. When set to "0" you're actually at +12dB. There is an internal gain structure that involves gains and losses all through the desk. You're never at "unity gain". Ever. There's always gain somewhere for it just to work as a mixer.
Thus, the target of a sweet spot at unity gain is a myth. The gain structure of a desk requres input calibration first, then all the other adjustments will work best if they are around their "0" position. But that's not unity gain.
How about you pad the output and move on?
No, I have to set you straight. Dynamic range reduction devices with adjustable attack times and ratios are known as compressor/limiters because they will do either task depending on how they are set. The topology of the detection method and the topology of the gain reduction circuit aren't especially defining either. True that optical compressors are slow to respond but the others not so. I've spent decades as a recording engineer and have used VCA, PWM, diode bridge and variable mu types. I own VCA types and used to own the great Manley Variable Mu compressor/limiter. I repair compressor/limiters as well. My Technology of Recording 465 class at MTSU in 1984 was clear about this. My experience over the years in the studio have agreed with what I was taught.
@jaddie, the fact that you think I'm "obsessed' shows you have some strange bias against my goal. I do appreciate your expertise though, so thank you for that but it's difficult to take you seriously when you make such outlandish statements.... I'm simply here to gain a better understanding and I have accomplished that so far. Not tryin' to throw shade your way.
You may object to me using analog but that is the scope of my question. Understanding the voltage going into the amp is very basic/core knowledge and not obsessive 😉 You learned at one time about this did you not? You sassy.
You may object to me using analog but that is the scope of my question. Understanding the voltage going into the amp is very basic/core knowledge and not obsessive 😉 You learned at one time about this did you not? You sassy.
I actually think we're in more agrement than you might think.
Let's try this:
You have a dyanmics processor., It has an infinite slope/ratio. It's threshold is set to -25dB re: system reference level. It uses an RMS detector, and it's attack time is 1dB per second, it's releaes time is 1dB/10 seconds. Limiter?
I appreciate and respect your credentials. We're not in competition.