Recording volume Ok. Playback volume too loud.

MilkoNetherland

New member
Dear recording enthusiasts.

I recently invested in some recording gear.
1xbehringer eurodesk 3242 fx-pro mixer
1x alesis HD24 HD recorder
3x Behringer protector mdx1800 8 channel limiter.

I use Y insertion cables from the mixer to the protector, Jack jack from the protector to the alesis and then from the output of the alesis back in the return of the Y cable back to the mixer. All pretty standard I would say.

Now the problem is that when I am recording I set the limiters so the alesis is using most of the 24 bit data so the volumes are never in the red but do get into the -1 area at peaks. This way I have maximum signal information into the recorder that is not distorted or peaked when I check the soundfile later and zoom in to the peaks.
But in playback mode the Alesis signal is too loud for the eurodesk mixer so teh mixer starts clipping like crazy..

Now the question arrizes... Does the alesis and the behringer have completely different standards for the 0db limit?
How can it be that a limited signal that is perfect for the recorder is heavily distored when it gets back into the mixer?

Is Behringer really that Shitty?? i assume and read that the alesis is build according to universal standards. But as far as I know in=out.. So there is no volume control in the alesis for the in or output. Am I missing something??

Now I can not use the mixer for playback unless I record very soft signals.

Please let me know if there is an solution to this problem.

Regards.

Milko
 
Yes, there ARE two different meter scales at work here but this time it's not a Behringer problem.

The Alesis HD24, being a digital device, has metres (and recordings) using the dBFS scale (dB Full Scale). I this metering, the zero level and clip level are one and the same, i.e. 0 on the Alesis meters is the maximum level that the recorder can handle.

Your mixer is an analogue device. Analogue meters are usually dBu where the zero level relates to a somewhat arbitrary analogue voltage. In this scale, zero is roughly 18dB below the clip level.

What this all means is that when your Alesis is outputting just below 0 on its meters, the level reaching the mixer is roughly +18dB on the analogue scale.

Normal advice is NOT to do digital recordings as hot as possible but, rather, to aim for an average of around -18dB(FS) on the digital meters, with peaks around -10dB(FS) This still leaves some headroom when you get to an analogue device. Note that, once you start mixing, you would need to lower levels anyway since, in the mix, levels are cumulative and as you add more tracks you'd soon get into clipping anyway.
 
Yes, there ARE two different meter scales at work here but this time it's not a Behringer problem.

The Alesis HD24, being a digital device, has metres (and recordings) using the dBFS scale (dB Full Scale). I this metering, the zero level and clip level are one and the same, i.e. 0 on the Alesis meters is the maximum level that the recorder can handle.

Your mixer is an analogue device. Analogue meters are usually dBu where the zero level relates to a somewhat arbitrary analogue voltage. In this scale, zero is roughly 18dB below the clip level.

What this all means is that when your Alesis is outputting just below 0 on its meters, the level reaching the mixer is roughly +18dB on the analogue scale.

Normal advice is NOT to do digital recordings as hot as possible but, rather, to aim for an average of around -18dB(FS) on the digital meters, with peaks around -10dB(FS) This still leaves some headroom when you get to an analogue device. Note that, once you start mixing, you would need to lower levels anyway since, in the mix, levels are cumulative and as you add more tracks you'd soon get into clipping anyway.

Ahh. So technically the direct outs of the eurodesks should not be used to fill up a digital recorder then??(at least not to the full scale) I love full fat signals that use the full 24 bit. The MDX limiters will make a very nice compressed sound that is fat, full and actually sound very dynamic afterwards in the mix becourse you use the full 24 bit depht in the recorder.. This is my trick to get very dynamic sounds.. So the limiters set to almost 0db(FS) contrubutes in a very direct and dynamic sound in stead of taking the dynamics out of the sound. World up side down but still working well for me.. Is there a solution? something like a -18 db trick for the output of the alesis?
 
Last edited:
Since the mixer direct outs made the signals but they're too hot on the return, does that mean you are boosting with the limiter?

If you like the sound because of it being limited, lower the threshold of the limiter don't boost. The recording will be as good- better actually, as you see falsely hyped record levels are problematic.

I.e. the moral of the story is -18 to 20 or so average, -6 or so peak... is our system's normal level.
 
Interesting that Bari says max in is +22dBu, Alesis +19dBu out.

Maybe the trim via line inputs can help.
 
Since the mixer direct outs made the signals but they're too hot on the return, does that mean you are boosting with the limiter?

If you like the sound because of it being limited, lower the threshold of the limiter don't boost. The recording will be as good- better actually, as you see falsely hyped record levels are problematic.

I.e. the moral of the story is -18 to 20 or so average, -6 or so peak... is our system's normal level.

Hi there.. No the limiters are not boosting anything.. They only limit to an adjustable level. In = out except if its louder than the set level. To use full capacity of my 24 bit i need to be just under the red in the alesis. So I setup my limiters and gain from the mixer so they only limit the loudest peaks and end up in the alesis at -1db. In any case I need to turn up volume louder than the mixers own return inputs can handle if i want to use the full bitdepht of the recorder. Even if I bypass the limiters the same thing happens.
Is there a -18db easy mod for signals? Stupid stuff that there are 2 standards at work...
 
While converters can have different alignments -as to where 0dBFS lands on the voltage in or out, yours is not a 'two standards' problem. Look here

http://hux.com.au/Soapbox%20Items/World%20Audio%20level%20Reference.pdf

What you are doing is deciding to not work within the 'nominal or sweet spot on your system.

It's one thing to like the sound of the limiting, or the perhaps stressed' analog parts of your chain but... Converter resolution (in your context) is defined', (right word?) by the ratio/level of the conversion error down below the recorded signal.
The converter's sound is the same as long as the error signals are well below your noise floor.
Almost guaranteed, the mixer, pre amps, noise floor 'defines' the limit of your 'quality'.. by about 20dB.
Running hot is basically just buying you... well, this situation.
Get twenty four line level attenuators?
 
While converters can have different alignments -as to where 0dBFS lands on the voltage in or out, yours is not a 'two standards' problem. Look here

http://hux.com.au/Soapbox%20Items/World%20Audio%20level%20Reference.pdf

What you are doing is deciding to not work within the 'nominal or sweet spot on your system.

It's one thing to like the sound of the limiting, or the perhaps stressed' analog parts of your chain but... Converter resolution (in your context) is defined', (right word?) by the ratio/level of the conversion error down below the recorded signal.
The converter's sound is the same as long as the error signals are well below your noise floor.
Almost guaranteed, the mixer, pre amps, noise floor 'defines' the limit of your 'quality'.. by about 20dB.
Running hot is basically just buying you... well, this situation.
Get twenty four line level attenuators?


Its also buying me more headroom in the recording because everything is louder so also low level sounds become brighter. Low level signals use just only a part of the 24 bit density(steps) so a soft recording around -18db uses only a part of the max possible information density compared to a signal that uses the full 0db scale and therefore the full 24 bit of information. This is why I get more dynamic end result when I run hot compared to running normal level.. Not many people use this technique but i swear it is working magically.. So the only problem I need to solve is the playback distortion caused by the loud signal... I will check the attenuators. Worst case I need to use 2 resistors and tap the line back to the mixer somewhere at 2/3rd.. In that case i need to know what impendance i can max use as load on the output of the alesis without killing the signal. 50kohm?? 100k ohm?? then all i need is 48 resistors..
 
Last edited:
' more headroom -because everything is louder- '
' low level sounds become brighter'
' Low level signals use just only a part of the 24 bit density(steps) so a soft recording around -18db uses only a part of the max possible information density'
(The man has- apparently, never made/experienced beautiful sounding tracks recorded -30 dBFS.)
' more dynamic end result when I run hot compared to running normal level'
' it is working magically'

:rolleyes:
You've reinvented how (in your mind that is..) digital conversion works. ;)
 
First, you need to get rid of the idea that you have to "fill up all 24 bits". The level of a sustained note should be somewhere around -18dbfs on the recorders meters. The rest of the room is for spikes and transients.

If you are using a limiter to stop the signal from clipping the recorder, you are changing the signal that you are recording (some might say ruining it), just so you can light up all the lights on the meter. That's a silly reason to permanently change the sound you are recording.

If you get rid of the limiters and adjust the recording level to the proper levels, all of your distortion problems will go away.

The way you have it set now, the gain staging is all messed up. That is causing the problems.
 
First, you need to get rid of the idea that you have to "fill up all 24 bits". The level of a sustained note should be somewhere around -18dbfs on the recorders meters. The rest of the room is for spikes and transients.

If you are using a limiter to stop the signal from clipping the recorder, you are changing the signal that you are recording (some might say ruining it), just so you can light up all the lights on the meter. That's a silly reason to permanently change the sound you are recording.

If you get rid of the limiters and adjust the recording level to the proper levels, all of your distortion problems will go away.

The way you have it set now, the gain staging is all messed up. That is causing the problems.

Technically and theoretically I understand exactly why you say this. And 10 years ago without having heared what I have heared now I would completely agree with you. The fact is that the recordings we make this way are used by Hifi extremists like Acapella CEO Rudolph to demonstrate his systems. There is a group named Hypnoclan that has some super dynamic sounding recordings that are all recorded with the same setup and input settings. Only difference is that the recorder we used back then did have an adjustable output... The kind of compression we achieve using this methode gives practically normalized tracks. Also it causes the low volume background sounds to come foreward in the sound image.. And this is achieved without giving the recording a compressed sound. I will share some of these tracks to prove this point. And using the full dynamic range in the 24 bit really isnt that silly.. Don't forget that i set the limiters to only come in in extreme spikes in the signal. And thats why I can just have a little fatter signal.

The mixer problem i have will be solved with attenuators. I calculated the -12 db would work for me. And a resistor of 5.1k and 1.3 k will probably do the trick. ill experiment on that,

Thanx folks so far.
 
' more headroom -because everything is louder- '
' low level sounds become brighter'
' Low level signals use just only a part of the 24 bit density(steps) so a soft recording around -18db uses only a part of the max possible information density'
(The man has- apparently, never made/experienced beautiful sounding tracks recorded -30 dBFS.)
' more dynamic end result when I run hot compared to running normal level'
' it is working magically'

:rolleyes:
You've reinvented how (in your mind that is..) digital conversion works. ;)

Yes that is true.. I never heard a good end result with low level recordings. Especially not when going -30.. What ever i try in EQ or compressors/limiters digitally afterwards or leaving is all original. It never sounds powerfull or any dynamic to me. Even with good mic's and professional recording rooms.. For me i know how I get the best results. And I will post a link to a soundcloud soon that is recorded by the person who introduced me to the hot recording technique. Really my only problem is the playback after the recording in my mixer. As soon I have it al in the computer I make a awesome mix of it.
Cheers!
 
Think about what you are doing...

You are using limiters, so that you can record about 12db louder...and then building an attenuator to keep you super loud recording from overloading the mixer channels.

All of the gear you are using is designed to work at line level. Running it way higher than it was intended to work is silly.
 
Think about what you are doing...

You are using limiters, so that you can record about 12db louder...and then building an attenuator to keep you super loud recording from overloading the mixer channels.

All of the gear you are using is designed to work at line level. Running it way higher than it was intended to work is silly.

The attenuators will make the mixer wort at line level again. So that problem is tackled. And was the cause of my question. I still need to upload oneof the hypnoclans songs to show of how dynamic the end results can be this way. Cheers
 
So, you actually think that the people who designed these products were stupid, and designed them so that they wouldn't work together very well... Of course, until you came along, figured out the riddle and outsmarted them...

Yup, that's probably what happened.


Something with large dynamics happens when the difference between the loudest part and the quietest part is really big. The reality is that 16 bits gives you more than enough dynamic range for any listenable music. The extra 8 bits in a 24 bit file just pushes the digital noise floor down below the noise floor of the rest of the system. The music will have just as much dynamics if you turn off the limiters and set the peaks to hit -6dbfs. The only difference is the digital noise floor will be at -118dbfs instead of -124dbfs. Of course, the noise floor of anything you record will.struggle to get below -96dbfs, so it doesn't make a difference.
 
MilkoNetherland, 24 bit recording doesn't give you more headroom (space at the top before distortion) or resolution (finer representation of loudness differences), it gives you more legroom (space at the bottom above the noise floor). No matter the number of bits or what level the signal, each bit equals the same ratio of loudness. Recording at -18dBFS average levels with 6dB or more between your peaks and 0dBFS gives you exactly the same resolution as slamming the signal into 0dBFS but without the distortion or need for limiters or overloading of downstream devices.
 
You both sort of said well, what I was thinking' -but kind'a gave up on, and or didn't care to get into- again.

There's another angle to look at some of the 'things we in this gig' do.. That is so many tracks, wonderful tracks, that get done sometimes in spite of what we do :rolleyes: or what we know (or don't believe... :>)
That says a lot to just how wide, clean, easy to get great sounds with such cheep (relative) tools we have now.
 
Last edited:
And furthermore, recording that hot means running your analog stuff on the input side above its optimal levels. They're designed to sound best (cleanest) with 0dBVU levels on the meter, which gives you +4dBu signal at the outputs which gives you (depending on converter calibration) -18dBFS in digital. Some preamps stay clean pushed to +8dBVU, some give you pleasant distortion (which you can still get at +4dBu if it has an output trim control) and some sound like ass.
 
I'm at a loss as to why you are even using limiters in the chain? There's no point. The digital dynamic range means all of it is usable - so all I do is make sure the preamps don't peak and just record it. Looking at my meters as long as the level is high enough to get the meters moving, I really pay little attention to it. I certainly haven't reduced my dynamics just to get stuff recorded? I have a pile of compressor limiters from my analogue days and they don't get used with any compression added after I have recorded, never before where it can't be undone!
 
That is the funny part about what he is doing...He is using a limiter to reduce the dynamic range of the signal, so that it uses all the bits, because he thinks that will get him more dynamic range.

It makes no sense, it makes it harder, and it adds needless things to the signal chain (which can only degrade the original signal)
 
Back
Top