The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
A thread for asking the things you should know by now but don't Audio Interfaces
Old 5th July 2018
  #5911
Lives for gear
 
12ax7's Avatar
 

Quote:
Originally Posted by YourBestFriend View Post
Should I ignore the clipping peaks and only pay attention to the rms levels? Keeping rms above -10 and then using a limiter to remove the clipping stuff?
You might get away with doing that (sometimes), but its not only poor practice, but also COMPLETELY UNNECESSARY!

In fact, there's absolutely no advantage at all to pushing levels anywhere near that high (until the mastering stage).

Here's a guide:
.
Old 5th July 2018
  #5912
Gear Head
 

ok..put my toys back in my pram and just spent the last few hours searching thru gearslutz for info about my original question...quite a few people have asked it before but none that i could find received a satisfactory answer...and its amazing how often the question is misunderstood and derailed with everyone getting lost in definitions...now i know why i never wanted to ask about it.

this one kinda got close....asked the right question.
Operating level calibration to monitor mastered material -XdbFS = ?

but i want to get to the bottom of it so..

it seems the hypothetical question should be about optimal settings for two channels of an apogee ad16x knowing it will always be fed 24bit, line level, modern loud, hyper mastered music that is already slamming near zero in the digital scale..there is no attenuation possible between the source file and the AD...it is loud and already full scale, no bits left maxed out etc...and you want to re-record as direct a copy of it as possible also peaking at -0.3 through the mastering pair of AD's with no further processing available...knowing this, what would the best reference level for my AD16x be.....??????

For those of you who are unaware..there is an option within the apogee ad16x that allows you to input a reference level to ensure you don't clip the ADC when recording...i personally use the setting of -18 when recording live bands through an ssl console, this works fine for me...and my own stereo mixes usually come in peaking between -6 and -3 on my daws digital full scale meters....then sent for mastering

but i am trying to determine if there is any sense in changing the -18 value of my mastering input channels to a lower value ie -3 when faced with the hypothetical scenario above...which would be my own masters....these particular masters are intended to be played back on traditional modern systems such as ipads, phones, line outs etc...while i dont crush my own masters, they do get brickwall limited to -0.3 before hitting the AD16x

it seems like a really straight forward question for this place...but, is never answered (as far as i could find) and derails easily, with the thread starters eventually giving up...i really hope someone can provide some insight into the specific hypothetical question...again i appreciate the replys from earlier.
Old 5th July 2018
  #5913
Lives for gear
 
bogosort's Avatar
Quote:
Originally Posted by smartfreq View Post
For those of you who are unaware..there is an option within the apogee ad16x that allows you to input a reference level to ensure you don't clip the ADC when recording...i personally use the setting of -18 when recording live bands through an ssl console, this works fine for me...and my own stereo mixes usually come in peaking between -6 and -3 on my daws digital full scale meters....then sent for mastering
Calibration mode on the AD16x sets the digital reference level (dBFS). You send a 1.23 Vrms (+4 dBu), 1 kHz analog sine to the input and use CAL mode to set the corresponding digital reference, e.g., -18 dBFS. The idea here is to pick a digital reference level that hits the sweet spot between headroom above and noise floor below.

Quote:
but i am trying to determine if there is any sense in changing the -18 value of my mastering input channels to a lower value ie -3 when faced with the hypothetical scenario above...which would be my own masters....these particular masters are intended to be played back on traditional modern systems such as ipads, phones, line outs etc...while i dont crush my own masters, they do get brickwall limited to -0.3 before hitting the AD16x
Ok, so what would calibrating the AD16x inputs to -3 dBFS do? You'd essentially be telling the converter that 1.23 Vrms (analog) equals -3 dBFS (digital). This is probably not what you want!

Suppose you send your mix to an outboard 2-bus compressor. Let's say that the compressor output has an average voltage of 8 Vrms (+20 dBu), a reasonable value for a squashed mix. If you route this to your -3 dBFS calibrated inputs, my guess is that you'll get severe digital clipping. You told the Apogee that +4 dBu = -3 dBFS, which means that anything over 4 + 3 = +7 dBu is a full-scale signal, and your average signal level is some 13 dB higher!

Had you left the calibration at +4 dBu = -18 dBFS, then the digital clipping point (0 dBFS) would correspond to a +22 dBu analog signal. With +20 dBu coming out of the compressor, you'd still have 2 dB of digital headroom. Remember, with 24-bit recording there's no good reason to ride as close as possible to 0 dBFS on the way in: once the signal is digitized, you can always apply gain to get the final level you want. So if I were you, I'd calibrate the AD16x down to +4 dBu = -20 dBFS for the extra headroom.
Old 6th July 2018
  #5914
Gear Head
 

Thank you Bogosort, for the perfect examples and detailed explanation ...the penny dropped, it finally makes sense to my brain..thank you also 12ax7 and Richard Crowley....i will be having another more detailed look at my set up and possible configurations....
Old 6th July 2018
  #5915
Here for the gear
 

Quote:
Originally Posted by thismercifulfate View Post
If your interface has let’s say 16 input channels but only 8 built-in mic preamps plus an ADAT input and you want to record signal from 16 mics, even if your interface has a db25 line input, every single mic still needs to go through a proper microphone preamp. The signal coming out of a mic is at mic level, which is a very faint signal. The primary function of a preamp is to boost said signal without adding noise to ‘line level’. So a mic preamp is designed to receive mic level signal, a d output line level signal. A line input is designed to receive line level signal. If you bypass a preamp and send a mic signal to a line input, the level will be as much as -60db quieter than where you need it to be. And you could try boost the level digitally in your daw, but the amount of noise you’d get would make the track completely unusable. So in summary, you can’t skip the mic preamp! If you want to record 16 mics, you need 16 mic preamps. There’s no way around it.
Thanks a bunch! That helps clarify big time. I was just curious if there were DB25 "ins" if a mic preamp could be bypassed with an XLR snake directly. This all makes total sense. Thanks for the clarity!
Old 7th July 2018
  #5916
Lives for gear
 
12ax7's Avatar
 

Quote:
Originally Posted by smartfreq View Post
Thank you Bogosort, for the perfect examples and detailed explanation ...the penny dropped, it finally makes sense to my brain..thank you also 12ax7 and Richard Crowley....i will be having another more detailed look at my set up and possible configurations....
...Kinda funny to me:

I knew in my heart that you were not an idiot (which is why I bothered to try to explain it to you).

And I kinda thought that if I could somehow simplify the explanation of it, you might understand...

...Little did I know that the real problem was that the concept was so damn simple that you overestimated its complexity:

...Therefore, you never understood it until 'Bogosort' explained it in more technical terms!
.

Last edited by 12ax7; 7th July 2018 at 01:50 AM..
Old 7th July 2018
  #5917
Gear Head
 

haha...i know...you had actually told me the answer..i just couldnt see it
well thanks for your patience and faith in humanity
Old 7th July 2018
  #5918
Lives for gear
 
12ax7's Avatar
 

Quote:
Originally Posted by smartfreq View Post
[...] thanks for your patience and faith in humanity
Hey, thank YOU!

It seems to be rare (in today's world) that a person might actually stick with some discourse long enough to reach some understanding.
.
Old 10th July 2018
  #5919
Gear Maniac
 

I haven't used a real one, but I have been wondering lately exactly how one is supposed to use the Vertical/Lateral mode on a Fairchild 670. Do you give it a mid mic and a side mic and it decodes them for you?
Old 11th July 2018
  #5920
Lives for gear
 
12ax7's Avatar
 

Quote:
Originally Posted by shoepedals View Post
I haven't used a real one, but I have been wondering lately exactly how one is supposed to use the Vertical/Lateral mode on a Fairchild 670. Do you give it a mid mic and a side mic and it decodes them for you?
You can use it pretty much like any mid/side device:

If you feed it with left/right channels, it'll turn it into a mid/side pair of signals, and process them.

If you want to, you could also (in the alternative) feed it a m/s pair of mics, and use it that way. (But you'll have to change the input configuration.)

Its kinda like the difference between petroleum and hemp:

Anything you can do with a hydrocarbon you can also do with a carbohydrate.
.
Old 11th July 2018
  #5921
Lives for gear
 
Quetz's Avatar
I have a 70's Rotel amp driving a pair of AKG LSM-50s (auratone clones).
The speakers are rated at 50W each, and the amp is only rated at 35Wpc, but turning it up to 3 is already way too loud.

Likewise with the headphone output - I get the same volume at level 2 on the Rotel HP out as I do with my Focusrite level all the way up.
Why is this?

I read (or dreamed) about someone writing about capacitor outputs, which lead some of these old amps to be under-rated on real power output, but I couldn't even tell you how a transformer works let alone a capacitor.

That admission is embarrassing enough in itself

Couple more amp-inspired questions:

The freq response for the amp's Aux and Tuner Ins is 8Hz - 45KHz. But all the other stats such as THD, Damping Factor, Continuous Power Output are based only on the 20Hz-20KHz range.
Likewise with audio interfaces, they can 'do' 96 and 192KHz sample rates, but where are the stats for the ranges beyond 20-20K?
Are those figures all over the place? Why aren't they considered important enough to publish, or are they too bad to publish?

If I'm sending out a 96KHz signal to an amp that extends to 45KHz then theoretically to speakers that go to 50KHz, isn't that more or less the ideal?
Or is this still considered unnecessary?
Old 11th July 2018
  #5922
Lives for gear
 
Piedpiper's Avatar
Quote:
Originally Posted by Quetz View Post
I have a 70's Rotel amp driving a pair of AKG LSM-50s (auratone clones).
The speakers are rated at 50W each, and the amp is only rated at 35Wpc, but turning it up to 3 is already way too loud.

Likewise with the headphone output - I get the same volume at level 2 on the Rotel HP out as I do with my Focusrite level all the way up.
Why is this?

I read (or dreamed) about someone writing about capacitor outputs, which lead some of these old amps to be under-rated on real power output, but I couldn't even tell you how a transformer works let alone a capacitor.

That admission is embarrassing enough in itself

Couple more amp-inspired questions:

The freq response for the amp's Aux and Tuner Ins is 8Hz - 45KHz. But all the other stats such as THD, Damping Factor, Continuous Power Output are based only on the 20Hz-20KHz range.
Likewise with audio interfaces, they can 'do' 96 and 192KHz sample rates, but where are the stats for the ranges beyond 20-20K?
Are those figures all over the place? Why aren't they considered important enough to publish, or are they too bad to publish?

If I'm sending out a 96KHz signal to an amp that extends to 45KHz then theoretically to speakers that go to 50KHz, isn't that more or less the ideal?
Or is this still considered unnecessary?
There's a difference between power output limits, input sensitivity, and gain.

And yes, anything beyond 20 to 20k is generally considered irrelevant. The advantage of higher sampling rates has to do with the type of filtering one is able to use that then effects phase issues within that 20 to 20k range, as well as the ability to process the audio more accurately. It's not to do with hearing stuff at 30k. Most mics roll off above 12K or so anyway. Many people argue that even those issues are not important at this point in the evolution of the technology.
Old 11th July 2018
  #5923
Lives for gear
 
GreenNeedle's Avatar
 

Quote:
Originally Posted by Richard Crowley View Post
Absent any specific details, that doesn't make any sense to me.
What exactly does "calibrated at -18dBFS" mean here?

Please note that the proper term is: dBFS
The "d" is lower-case because it represents the fractional prefix "deci" meaning 1/10th.
The "B" is upper-case because it is the initial of Alexander Graham Bell after whom the unit of measure is named.
The "FS" is capitalized by convention as it is the reference for the dB measurement. Note that dB by itself represents only a ratio and not a specific level.

Decibel - Wikipedia
Decibel - Wikipedia
dBFS - Wikipedia
Awesome, i always thought the acronym was for digital below full scale! Didn’t even notice the upper and lower case stuff!
Old 12th July 2018
  #5924
Lives for gear
 
Quetz's Avatar
Quote:
Originally Posted by Piedpiper View Post
There's a difference between power output limits, input sensitivity, and gain.

And yes, anything beyond 20 to 20k is generally considered irrelevant. The advantage of higher sampling rates has to do with the type of filtering one is able to use that then effects phase issues within that 20 to 20k range, as well as the ability to process the audio more accurately. It's not to do with hearing stuff at 30k. Most mics roll off above 12K or so anyway. Many people argue that even those issues are not important at this point in the evolution of the technology.
Thanks Tim.

So using 96KHz sample rates isn't done because it extends the frequency range but because it allows us to push artifacts further beyond the 20K limit than a 44.1K samplerate would allow.
And in so doing it means that the 20-20K range is far 'cleaner' than it otherwise would be, hence the ability to process that range more accurately.
Have I paraphrased you correctly?

The following question might seem like baiting to some, but I assure you it's not, I just want to ascertain what is and what isn't worth worrying about:

I recently bought an EQ plug that allows me to boost at 40KHz (Brainworx v3 digital).
I can clearly hear that boost.
So am I hearing a boost at 40K, or am I hearing the impact that boost has on lower frequencies with common denominators, if that makes sense?
As in, when you boost at 80Hz, you know you will be affecting multiples of that frequency going upwards, but if you boost at 8K or 40K, do those boosts affect the lower harmonics as well?

Thanks again,

Q.
Old 12th July 2018
  #5925
Here for the gear
 

Quote:
Originally Posted by Quetz View Post
I recently bought an EQ plug that allows me to boost at 40KHz (Brainworx v3 digital).
I can clearly hear that boost.
So am I hearing a boost at 40K, or am I hearing the impact that boost has on lower frequencies with common denominators, if that makes sense?
As in, when you boost at 80Hz, you know you will be affecting multiples of that frequency going upwards, but if you boost at 8K or 40K, do those boosts affect the lower harmonics as well?
Q.
The frequency that you are boosting is the centre of the range of frequencies. It will boost frequencies either side of that centre point. The Q factor of an EQ determines the width of the bell (narrow or wide), assuming you are talking about a parametric EQ rather than say a shelf filter. The bandwidth is the range or frequencies which are 3dB less than the boost (or cut). So if you boost by 10dB the frequencies either side which are boosted by 7dB determine the bandwidth of the EQ.

With a Q of 1 the bandwidth is the same value as the centre frequency so for centre at 40kHz will be 40kHz wide (roughly 25kHz to 65kHz). A higher Q means a tighter notch and smaller bandwidth.

So even with a centre frequency of 40kHz if you had a significant boost or a low Q factor you are going to hear the effect of the boost in the audible range even where the EQ frequency selected is centred outside of the audible range.

Cheers
Old 12th July 2018
  #5926
Gear Addict
 
YourBestFriend's Avatar
Quote:
Originally Posted by Quetz View Post
I have a 70's Rotel amp driving a pair of AKG LSM-50s (auratone clones).
The speakers are rated at 50W each, and the amp is only rated at 35Wpc, but turning it up to 3 is already way too loud.

Likewise with the headphone output - I get the same volume at level 2 on the Rotel HP out as I do with my Focusrite level all the way up.
Why is this?

I read (or dreamed) about someone writing about capacitor outputs, which lead some of these old amps to be under-rated on real power output, but I couldn't even tell you how a transformer works let alone a capacitor.

That admission is embarrassing enough in itself

Couple more amp-inspired questions:

The freq response for the amp's Aux and Tuner Ins is 8Hz - 45KHz. But all the other stats such as THD, Damping Factor, Continuous Power Output are based only on the 20Hz-20KHz range.
Likewise with audio interfaces, they can 'do' 96 and 192KHz sample rates, but where are the stats for the ranges beyond 20-20K?
Are those figures all over the place? Why aren't they considered important enough to publish, or are they too bad to publish?

If I'm sending out a 96KHz signal to an amp that extends to 45KHz then theoretically to speakers that go to 50KHz, isn't that more or less the ideal?
Or is this still considered unnecessary?
I am about to buy an old silver face and I've been reading every forum I can find about old receivers. Lots of interesting information. Talk of how in the 70s the manufacturer was required to warm up the receiver for an hour before very specific testing for wattage.

I came across interesting ideas as to why turning up the receiver to max and the preamp down low would be preferred. The pot switches on the volume knobs open up wider to certain frequency when fully cranked, or something to that extent. The expensive power amps don't even come with volume knobs.

I was reading about unity gain, and the windows volume knob. People where saying below 30 percent the bit rate drops. There were arguments as to weather unity gain was at 50 or 100 percent.

I never ran my receivers like this when I was younger but I am interested in trying it out soon.
Old 12th July 2018
  #5927
Lives for gear
 
Quetz's Avatar
I was lucky enough to get the original owner's manual with my Rotel RA-350, and there is no mention of such behaviour at all, and the specs sheet on the back page is pretty comprehensive.
My dad had a Pioneer SA-508 or 608 blueline, think it was one of those two, I loved that thing when I was a kid.
I needed an amp and was looking for the same Pioneer, but I just missed all the decent priced auctions, so went for the RA-350, and I'm not disappointed.

For £45 you can't really complain
Old 12th July 2018
  #5928
Lives for gear
 
Quetz's Avatar
Quote:
Originally Posted by Perwinkle View Post
The frequency that you are boosting is the centre of the range of frequencies. It will boost frequencies either side of that centre point. The Q factor of an EQ determines the width of the bell (narrow or wide), assuming you are talking about a parametric EQ rather than say a shelf filter. The bandwidth is the range or frequencies which are 3dB less than the boost (or cut). So if you boost by 10dB the frequencies either side which are boosted by 7dB determine the bandwidth of the EQ.

With a Q of 1 the bandwidth is the same value as the centre frequency so for centre at 40kHz will be 40kHz wide (roughly 25kHz to 65kHz). A higher Q means a tighter notch and smaller bandwidth.

So even with a centre frequency of 40kHz if you had a significant boost or a low Q factor you are going to hear the effect of the boost in the audible range even where the EQ frequency selected is centred outside of the audible range.

Cheers
Thank you for the breakdown I was aware of how a bell curve worked/looked, but I didn't know how Q number related to bandwidth in the detail you've provided so thanks, that's really helpful and useful.

What I was trying to get at specifically, was behaviour beyond the boundaries of the eq curve.
The following is an excerpt from an article on my grotboxes, and it's what triggered the original question:

Quote:
Originally Posted by Ben Duncan - Home Recording Studio Magazine
..much depends on how smoothly the bass rolls off, and providing the bass harmonics are reproduced accurately your ears will synthesise the fundamental. That is, you'll hear much of the (physically) absent low bass, especially from a good, phase-coherent recording. The bass harmonics, of course, lie in the midrange zone, where excellent performance can be expected from a 5½" diameter cone. At the same time, the relatively gentle bass roll-off helps to maintain original phase relationships between any fundamental that is audible, and the upper harmonics.
So can we say that provided we're working with material with excellent phase coherency, that affecting a high frequency harmonic will have an effect on the way we perceive that sound's lower freq. harmonics/fundamental (or only the fundamental?), insofar as our ears will try to reproduce them, and would this apply to all frequencies, not just bass frequencies as described in the article?
Old 12th July 2018
  #5929
Gear Guru
 

Quote:
Originally Posted by Quetz View Post
And in so doing it means that the 20-20K range is far 'cleaner' than it otherwise would be
It takes a load off the filter. It's cleaner, but "far" cleaner is debatable. As Piedpiper said, the technology has evolved quite a bit. A lot of people -including many professionals - fail blindfold testing to discern material recorded at different sample rates. Of the ones that succeed, most of them are probably picking up on the differences in the filters, rather than, let's say "hearing" 40kHz.

Quote:
So am I hearing a boost at 40K, or am I hearing the impact that boost has on lower frequencies with common denominators, if that makes sense?


when you boost at frequency "x" you are also boosting above or below frequency X. Depending on the Q, more or less stuff is "included" in your boost. Totally likely that the curve of your boost extends down to the audible range. Totally unlikely that you can hear 40kHz.


Quote:
Originally Posted by Quetz View Post
Likewise with audio interfaces, they can 'do' 96 and 192KHz sample rates, but where are the stats for the ranges beyond 20-20K? Are those figures all over the place? Why aren't they considered important enough to publish
maybe because human beings cannot hear them? Even most of your high-dollar microphones drop off sharply above 20k. What microphones do you own that can capture significantly above 20kHz? What instruments (besides cymbals) are making significant amounts of sound in the 20k -45k region? A region that you can't hear anyway. So your signal has the potential to 'extend to 45kHz' - what's in there really? And why should you care?

Quote:
If I'm sending out a 96KHz signal to an amp that extends to 45KHz then theoretically to speakers that go to 50KHz, isn't that more or less the ideal?
You might ask yourself, why do I own this hot-rod sports car when the speed limit in my state is 65? Then you have to pass a slow moving truck on hill on a 2-lane blacktop, and you know the answer.
Old 12th July 2018
  #5930
Gear Addict
 
YourBestFriend's Avatar
Paying attention to peak level seems like a joke. Rms seems to be the only thing that matters. what would happen if i chose not to monitor peak levels, but only the rms?

Are the transients just bits of broken overly loud information?
Old 13th July 2018
  #5931
Lives for gear
 
12ax7's Avatar
 

Quote:
Originally Posted by YourBestFriend View Post
Paying attention to peak level seems like a joke. Rms seems to be the only thing that matters. what would happen if i chose not to monitor peak levels, but only the rms?

Are the transients just bits of broken overly loud information?
With all due respect:

To say that "Paying attention to peak level seems like a joke." is to ignore this fact:

Exceeding maximum (dBFS) peak level always results in the "chopping off" of the top of the peaks of the waveform.

To believe this doesn't matter is to believe that clipping doesn't change the sound (which is inherently absurd).

...And it gets even worse when considering your second statement:

"Transients [are] just bits of broken overly loud information".

This statement could only be true if the peaks have been clipped (as a result of ignoring peak level).

Futhermore:

If "RMS is the only thing that maters", then I guess the best recording ever would be a 3 minute track of pink noise (with the peaks chopped off).

Is it really true that the only thing that matters is the peak-to-average ratio?

REALLY??
.
Old 13th July 2018
  #5932
Gear Addict
 
YourBestFriend's Avatar
Quote:
Originally Posted by 12ax7 View Post
With all due respect:

To say that "Paying attention to peak level seems like a joke." is to ignore this fact:

Exceeding maximum (dBFS) peak level always results in the "chopping off" of the top of the peaks of the waveform.

In light of this, to believe that ignoring peak level makes no difference is to believe that clipping doesn't change the sound (which is absurd).

It gets even worse when considering your second statement:

"Transients [are] just bits of broken overly loud information".

This statement could only be true if the peaks had not been clipped as a result of ignoring peak level.

Futhermore:

If "RMS is the only thing that maters", then I guess the best recording ever would be a 3 minute track of pink noise (with the peaks chopped off).

Is it really true that the only thing that matters is the peak-to-average ratio?

REALLY??
.
the second question wasn't a statement. it seems I'm asking what are, and why deal with, the transients? are the transients just higher octaves of say a cymbal crash? not super educated or trying to make statements. just questions.

it seems playing a bassline at -15rms with the peaks near zero is powerless compared to a bassline at -8rms with the peaks getting smashed. i feel like im not feeding the master buss enough juice when im not clipping.

in the analogue days was clipping less avoided?
Old 13th July 2018
  #5933
Lives for gear
 
12ax7's Avatar
 

Quote:
Originally Posted by YourBestFriend View Post
[...]
in the analogue days was clipping less avoided?
You're damn right it was!

(It was avoided like the plague!)

Everybody knew that when they sent their stuff to tape, it was gonna get "saturated", and then again transferred to yet another track (or two), creating even more distortion.

...Also, we also knew that the resulting tape was then again going to be further distorted upon transfer to disc (and then again further degraded upon pressing to vinyl).

We were trying to FIGHT distortion (except for a choice of effect upon a certain instrument).

Clipping was something that only happened when we wanted to get a guitar sound or some other "artistic choice" (almost always done on a track-by-track basis).
.

Last edited by 12ax7; 13th July 2018 at 12:49 AM..
Old 13th July 2018
  #5934
Gear Addict
 
YourBestFriend's Avatar
Quote:
Originally Posted by 12ax7 View Post
You're damn right it was!

(It was avoided like the plague!)

Everybody knew that when they sent their stuff to tape, it was gonna get "saturated", and then mixed again to tape (creating even more distortion)

...And then we also knew that it was then again going to be further distorted upon transfer to disc (and then again further degraded upon pressing to vinyl).

We were trying to FIGHT distortion (except for a choice of effect upon a certain instrument).

Clipping was something that only happened when we wanted to get a guitar sound or some other "artistic choice" (almost always done on a track-by-track basis).
.
Were rms levels ignored? did you just aim for as close to zero with the peaks? is there a way to reduce transients without a compressor? can a singer have less transients in their voice?
Old 13th July 2018
  #5935
Lives for gear
 
12ax7's Avatar
 

Quote:
Originally Posted by YourBestFriend View Post

Are rms levels ignored?
Sure. (Never even thought about it)

If we wanted stuff louder, we'd just grab the volume knob, and turn it clockwise.

This ALWAYS results in an increase in RMS. (There's no real trick to it.)
Quote:
Originally Posted by YourBestFriend View Post

did you just aim for as close to zero with the peaks?
We seldom even looked at peaks.

(Hell, we seldom even had any meter that would allow us to do that).

We just listened.

(Analog tape capture doesn't "crash" when it hits a "brick wall" of 0dBFS like digital does.)
Quote:
Originally Posted by YourBestFriend View Post

is there a way to reduce transients without a compressor?
Sure there is! (There are plenty of ways to do that.)
Quote:
Originally Posted by YourBestFriend View Post

can a singer have less transients in there voice?
It comes to mind that maybe we need to define our terms here:

What do you mean by "transients"?
(I'm starting to think that a lotta folks don't even seem to know what that term really means.)
Old 13th July 2018
  #5936
Gear Guru
 

Quote:
Originally Posted by 12ax7 View Post
[INDENT]
We seldom even looked at peaks.

(Hell, we seldom even had any meter that would allow us to do that).
A few of my decks had an overload light associated with the meter. There was almost no way of recording without lighting up those lights, so our goal was to just not have them light up Like Crazy.

Quote:
It comes to mind that maybe we need to define our terms here:

What do you mean by "transients"?

(I'm starting to think that a lotta folks don't even seem to know what that term really means.)
the popular Gearslutz definition is: "those evil qualities of real-life sounds that have to be killed for recording purposes"

...so that we can have higher and higher averages, I guess!
Old 13th July 2018
  #5937
Lives for gear
 
12ax7's Avatar
 

Quote:
Originally Posted by joeq View Post
A few of my decks had an overload light associated with the meter. There was almost no way of recording without lighting up those lights, so our goal was to just not have them light up Like Crazy.



the popular Gearslutz definition is: "those evil qualities of real-life sounds that have to be killed for recording purposes"

...so that we can have higher and higher averages, I guess!
As far as I can gather, the general idea is it make sure you pay lots of money so you can have a mic with low distortion, a very wide frequency response, and good transient response.

...That way you can use a high-pass filter plugin to get rid of the low end, a low-pass filter plugin to get rid of the high end, a saturation plugin to generate distortion and a brickwall limiter to eliminate the transient peaks.

PERFECTION!!!
.
.
Old 13th July 2018
  #5938
Lives for gear
 
12ax7's Avatar
 

Quote:
Originally Posted by joeq View Post
the popular Gearslutz definition is: "those evil qualities of real-life sounds that have to be killed for recording purposes"

...so that we can have higher and higher averages, I guess!
As far as I can gather, the general idea is to make sure you pay lots of money for a mic with very wide frequency response, low distortion, and good transient response...

...That way you can use a high-pass filter plugin to get rid of the low end, a low-pass filter plugin to get rid of the high end, a saturation plugin to generate distortion, and a brickwall limiter plugin to eliminate the transient peaks.

When you're done with all that, you put the result through a series of plugins to increase the RMS, and then put on a final limiter plugin (or two).
(...Then you sit back and complain about how much better it was when everything was analog.)
.

Last edited by 12ax7; 13th July 2018 at 02:22 AM..
Old 13th July 2018
  #5939
Lives for gear
 
Quetz's Avatar
Thanks for the response

Quote:
Originally Posted by joeq View Post
Of the ones that succeed, most of them are probably picking up on the differences in the filters, rather than, let's say "hearing" 40kHz.
I was just reading in the Geoff Emerick Q&A that story about him and the Neve desk, where after going through various processes Mr. Neve and himself established he was 'feeling' harmonics "derived from 50K".
I'm not even sure what that means.

Quote:
Originally Posted by joeq View Post
Even most of your high-dollar microphones drop off sharply above 20k. What microphones do you own that can capture significantly above 20kHz?
I always forget about the microphones, as if we could just pluck these sounds out of the air.
And, by the way, why can't we do that yet?
Is anybody even trying?

We need a different paradigm to the microphone.

Quote:
Originally Posted by joeq View Post
What instruments (besides cymbals) are making significant amounts of sound in the 20k -45k region? A region that you can't hear anyway. So your signal has the potential to 'extend to 45kHz' - what's in there really? And why should you care?
I care because I believe that god is in the details, as it were.
I have had many experiences in my life where I've learned to place just as much importance on what I feel as what I see/hear.
And so I don't see it as being a case of 'what you can't really hear', it all has an impact on the overall experience, probably more than we think.

Just because it's not loud in those frequencies doesn't mean they aren't important. People don't listen to music like robots, generally. We like music because it makes us feel, and granted someone listening to end product on ear buds isn't shoring up the argument that hi-fidelity is the only thing that makes music emotionally stimulating, but the focus here is on the creators and the engineers whose moment-by-moment emotional decisions are influenced by fidelity.
If music wasn't recorded, produced and mixed emotionally, then they would have made robots/machines that could do it automatically.

I'm out on a limb here I know, and this is more philosophy than technical..
Old 14th July 2018
  #5940
Gear Guru
 

Quote:
Originally Posted by Quetz View Post
Mr. Neve and himself established he was 'feeling' harmonics "derived from 50K". I'm not even sure what that means.
I have read the story too, and I am not sure what it means either, but some people around here take the message that a module was malfunctioning at 50k and Mr Neve could hear there was something wrong with it. and therefore Mr Neve's hearing goes to 50k! Since that is well more than twice as high as any human being has ever demonstrated the ability to hear, since that would be Ultrasonic to a dog, I take a different view - which is that it is entirely possible that a module that is malfunctioning at 50k may also be producing subtle sounds in the audible range at the same time.

Quote:
And so I don't see it as being a case of 'what you can't really hear', it all has an impact on the overall experience, probably more than we think.
There are already plenty of threads on this topic if you are interested. I think science has already told us what the reality of what we can hear and cannot hear is. Experiments have been done where ultrasonic content of the music was removed and put back removed and put back. Not only could nobody tell by guessing, nobody 'liked' or 'disliked' the music any more, nobody's brain scans registered any differently, . No conscious, subconscious or neurological reaction has been shown to take place. What more do you want? There are some good reasons to build overkill/headroom into our gear, but magical enhancement of the listening experience via ultrasound is probably not one of them.

Quote:
Just because it's not loud in those frequencies doesn't mean they aren't important.
I can only tell you that I have already heard these ideas and considered them well, and personally I am not buying it. There is no place on the tonotopic map of the basilar membrane where sounds higher than 20k can even focus.



The location of these 'sensors' is well-studied and damage to a specific section is consistent with hearing loss at those frequencies. Where's the 50k section?

I suppose if you focused a powerful enough ultrasonic beam at someone you could heat up their flesh and destroy tissue. That's not a "form of hearing" by my definition, and in any case, I doubt your stereo is up to the job.

Quote:
I'm out on a limb here I know, and this is more philosophy than technical..
"Ultrasonic hearing" is notoriously the last refuge of Audiophools. They push it hard because then they can use it to justify pretty much anything. In my view, the only thing that could overthrow my confidence in what science currently says about ultrasonic hearing (which by definition is hearing sound you can't hear! ), would be new scientific discoveries.

Not "philosophy". Not clever arguments. Not anecdotes.

If there ever are new scientific proofs that humans can hear the sounds that humans cannot hear, I am sure we here at GS will be among the first to learn about them. If they are good studies, with real scientific methodology, and stand up to replication, I will change my mind.
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Similar Threads
Thread
Thread Starter / Forum
Replies
Jim Roberts / The Good News Channel
7
drockfresh / The Good News Channel
10
David Herbert / The Good News Channel
31
tnjazz / Gearslutz Secondhand Gear Classifieds
0

Forum Jump
Forum Jump