When you listen to live music, for example a drum kit, the sound levels at times can exceed 100dB close up. Orchestras are even louder. The power of live music is one feature of live music, there is the sheer smoothness or linearity or fluidity of the live voice or instrument, missing the granularity I hear or imagine I hear from music reproduction.
Now playing music in the average living room I will not or cannot play back at such high sound levels, or maybe at the (guessed) mixing levels. So at one go, the accuracy of the reproduction is diminished. It is like listening to a band playing in the distance but that also introduces echo and wind effects.
What I hear then is a miniature version of the recorded music, even if I compensate for loudness effects.
Aren't we listening to miniatures of music rather than music? How to equalize this so it resembles as far as possible what the live listener hears or recording studio monitors put out?
As I delve deeper into building speaker cabinets and testing micro power amplifiers, these questions seem to spring up from time to time.
Now playing music in the average living room I will not or cannot play back at such high sound levels, or maybe at the (guessed) mixing levels. So at one go, the accuracy of the reproduction is diminished. It is like listening to a band playing in the distance but that also introduces echo and wind effects.
What I hear then is a miniature version of the recorded music, even if I compensate for loudness effects.
Aren't we listening to miniatures of music rather than music? How to equalize this so it resembles as far as possible what the live listener hears or recording studio monitors put out?
As I delve deeper into building speaker cabinets and testing micro power amplifiers, these questions seem to spring up from time to time.
Last edited:
Re: How to equalize this so it resembles as far as possible what the live listener hears or recording studio monitors put out?
Great question! As a ten year ex recording/mixing engineer, perhaps this will assist.
Recording studio control rooms, are usually set up to meet industry guidelines, at least the good ones are. Here are a couple of examples of those guidelines:
Recommendation ITU-R BS.1116-3 (02/2015) Methods for the subjective assessment of small impairments in audio systems.
Listening conditions for the assessment of sound programme material: monophonic and two–channel stereophonic.
There are several others, but these two, are not only similar, but so are the others. If you spend the time reading them, you will find out that we are talking “stereo” here, which is an illusion, a good illusion, but not what the live listener hears, but what the recordings studio monitors put out.
So to replicate, as close as possible, one will want to set up your music playback system similarly. For example, in a 60 degree equilateral triangle. With the speakers frequency response, shaped the same way as in the guidelines. There are several other specs, like speaker directivity, room reverb times, etc., that will help one get closer to similar operating conditions that were in the studio control room.
This includes the ears non-linearity to amplitude versus frequency response. Most monitoring engineers will mix the music at reference level (i.e. 83 dB SPL) because the ear is at its (relative) flattest response. This is a really good article on the subject of what level to monitor at: Level Practices (Part 2) - Digido.com If listening at lower levels, then use a dynamic loudness control like in JRiver’s Media Center software player.
If we want accurate music reproduction, then the best we can hope for is that the music arriving at ones ears over loudspeakers matches as close as possible to what is recorded on music media. This means the frequency response is smooth and matches one of the recommended target responses and the direct sound is arriving all at the same time with minimal phase distortion. The only way to achieve this currently in the home environment is to use DSP for loudspeakers in rooms.
Here is an article I wrote recently that shows what is possible: Audiolense Digital Loudspeaker and Room Correction Software Walkthrough - CA Academy - Computer Audiophile
This is a huge subject area and near impossible to properly represent in a forum thread. If you want to pursue further beyond the articles listed, I wrote a book on the subject in my sig. If you click on Look Inside, you can peruse the table of contents and read the first few chapters for free to see if it is for you or not.
Personally, using modern DSP software to correct both the frequency and timing response in my room, gets me much closer to the experience I had at the studio at a fraction of the cost. At reference level, the recordings sound full, large and if I turn it up, I can play along with my acoustic drum kit in the same room as my stereo. Lots of fun.
Good luck on your journey!
Great question! As a ten year ex recording/mixing engineer, perhaps this will assist.
Recording studio control rooms, are usually set up to meet industry guidelines, at least the good ones are. Here are a couple of examples of those guidelines:
Recommendation ITU-R BS.1116-3 (02/2015) Methods for the subjective assessment of small impairments in audio systems.
Listening conditions for the assessment of sound programme material: monophonic and two–channel stereophonic.
There are several others, but these two, are not only similar, but so are the others. If you spend the time reading them, you will find out that we are talking “stereo” here, which is an illusion, a good illusion, but not what the live listener hears, but what the recordings studio monitors put out.
So to replicate, as close as possible, one will want to set up your music playback system similarly. For example, in a 60 degree equilateral triangle. With the speakers frequency response, shaped the same way as in the guidelines. There are several other specs, like speaker directivity, room reverb times, etc., that will help one get closer to similar operating conditions that were in the studio control room.
This includes the ears non-linearity to amplitude versus frequency response. Most monitoring engineers will mix the music at reference level (i.e. 83 dB SPL) because the ear is at its (relative) flattest response. This is a really good article on the subject of what level to monitor at: Level Practices (Part 2) - Digido.com If listening at lower levels, then use a dynamic loudness control like in JRiver’s Media Center software player.
If we want accurate music reproduction, then the best we can hope for is that the music arriving at ones ears over loudspeakers matches as close as possible to what is recorded on music media. This means the frequency response is smooth and matches one of the recommended target responses and the direct sound is arriving all at the same time with minimal phase distortion. The only way to achieve this currently in the home environment is to use DSP for loudspeakers in rooms.
Here is an article I wrote recently that shows what is possible: Audiolense Digital Loudspeaker and Room Correction Software Walkthrough - CA Academy - Computer Audiophile
This is a huge subject area and near impossible to properly represent in a forum thread. If you want to pursue further beyond the articles listed, I wrote a book on the subject in my sig. If you click on Look Inside, you can peruse the table of contents and read the first few chapters for free to see if it is for you or not.
Personally, using modern DSP software to correct both the frequency and timing response in my room, gets me much closer to the experience I had at the studio at a fraction of the cost. At reference level, the recordings sound full, large and if I turn it up, I can play along with my acoustic drum kit in the same room as my stereo. Lots of fun.
Good luck on your journey!
I highly doubt that MOST recordings were mixed at that LOW of a listening level. In my years working in recording studios, I NEVER heard one mix done at that level.
I haven't been in a studio in a few years but what I remember is that music mixes were loud. The only standard I saw was make it loud enough to impress the band, which isn't easy when there used to standing in front of guitar amps and live drums. Which is too bad because mixes would be more consistent if there was a standard lower level. Where I've seen a standard is in audio for video (tv and movies). Dialogue normal is 86dbspl at 0VU for tvs and 92dbspl for movies. And then for tv we would listen to playback over a mono 3" crap speaker at living room levels.
It’s intresting to me that we are talking about reproducing a “Live” experience as if there is nothing between us and instruments or performers. I love to go to live performances, and my favorites are more intimate venues like jazz clubs. Unfortunately, I have been to see favorite artists at these venues where one would think you would bask in the unique “live” sound quality only to be disappointed and experiencing what sounds like an over engineered performance or a recording. I leave confused, only to appreciate a gent playing acoustic guitar in th subway
Fritz Langford Smith in the audio section of his Radio Designer's Handbook said that the object of music reproduction was to produce the same emotions ,as closely as possible,in the listener as hearing a live performance.Obviously,it is not possible to have a live band,or a full orchestra in one's lounge,so we have to settle for electronic sound reproduction!
lol I think you need more ALL CAPs as we can't hear you
Unfortunately, one of the biggest myths in the industry... and the reason we have industry guidelines. Did you read any of the ones linked? All state what the reference listening level should be. How about Bob Katz's the magic of 83 db SPL?
Anyway, you will find many references to "reference" listening level - it's an industry standard. Why? Science folks.
Our ears frequency response changes with SPL. At low SPL, like 65 db SPL, our ears don't hear the bass very well relative to other frequencies. That's why there is the science of equal loudness contours. As the SPL goes up, our ears hear realtiveley flat in the bass at around 83 dB SPL. Now as we increase the SPL to say 100 dB SPL, our ears sensitivity to bass goes up. This is important to keep in mind because...
Ever listen to early rock bands and wonder why the bass sounds so thin on recordings? That's because they were mixed too loud. Remember our ears bass response changes with level. So when mixing rock at 100 or 105 db SPL (i.e. max reference level), the mixer has a tendency to turn the bass down as the ear is more sensitive at this level and sounds out of balance in the mix relative to other frequencies. However, when played back at a normal listening level, like reference level for example, the mix sounds thin, because the ear is less sensitive at this SPL. Totally counter intuitive for the folks that never went to an audio engineering school or did not read and understand the science.
So if you mixed it at reference and then turned it way up to listen, and impress anyone around you, the bass sounds thunderous, like it should
Now mixing has gone to the opposite extreme where mixers are listening to as low as a level that they can get away with, like around 77 dB SPL and pump up the bass and the loudness so that the recording stands out, sounding "full blast", even at a low SPL. This is called the loudness wars and I wrote an article about it here: Dynamic Range: No Quiet = No Loud - CA Academy - Computer Audiophile
Anyway, best to look at industry guidelines and the science behind (i.e. equal loudness contours) as to why pros mix at around ~83 dB SPL (see Bob Katz above) and why we as listeners, who are trying to hear as accurately as possible, monitor around the same level. Because xkcd: Science
Enjoy the music!
While I read your replies, here is what I am talking about,
If a live band was reduced to this size, what would it sound like? Or what should it sound like?
Lower SPL but loudness compensated at the higher and lower ends? How does reverberation change with sound levels? Stereo effects?
I guess what I am asking is can the Hi-Fidelity experience happen at low SPLs or should we be content the same way we are content watching a football game on a 3 inch screen?
If a live band was reduced to this size, what would it sound like? Or what should it sound like?
Lower SPL but loudness compensated at the higher and lower ends? How does reverberation change with sound levels? Stereo effects?
I guess what I am asking is can the Hi-Fidelity experience happen at low SPLs or should we be content the same way we are content watching a football game on a 3 inch screen?
Well not really, how would this be done? 1 metre distance from speakers and they have to be monitor type speakers?
Barcus Berry Electronics (BBE) is a digital enhancememt to recordings to help recreate the presence of a live performance.
The piano notes sound sharper, the strings have more detail in their vibrations.
I just wonder if its not the power of the notes that you seek but rather its attitude or rise time or attack curve.
The piano notes sound sharper, the strings have more detail in their vibrations.
I just wonder if its not the power of the notes that you seek but rather its attitude or rise time or attack curve.
I think I know what you are getting at. I have full range speakers and subs. At low SPL, loudness compensated, (I use this), they still sound like full range speakers and subs, with the same mix, reverb, stereo effects, etc., but simply quieter. It does not sound like coming through a 3 inch speaker. It still sounds high fidelity. That is if I get the drift of your question...
If a band was reduced to a few inches in height and the instruments they played were in scale then the sound they would produce would be lacking in low/bass tones.
The spectrum of frequencies we hear declines near the lower frequencies so a big drum or speaker is needed to make those sounds more prominent.
Because the bass sounds are lacking in volume then the higher pitched sounds will be more prominent and even dominate the music spectrum.
A 6 inch speaker can.. for example produce the rumbling bass note of a bass drumkit however not at a sufgiciently high volume to hear it properly.
So... your tiny beatles band would sound tinny and lacking in bass.
The spectrum of frequencies we hear declines near the lower frequencies so a big drum or speaker is needed to make those sounds more prominent.
Because the bass sounds are lacking in volume then the higher pitched sounds will be more prominent and even dominate the music spectrum.
A 6 inch speaker can.. for example produce the rumbling bass note of a bass drumkit however not at a sufgiciently high volume to hear it properly.
So... your tiny beatles band would sound tinny and lacking in bass.
Yes this is what I mean, really, hard to match the attack curve of a loud thump on a table for example.
Ah yes this is what I want. Could you post a note about your set up? Says a lot for full range, what I am using now, on a small scale.
Yes this is what I am getting at, need to compensate for sound, although the possibility of a small band like this is remote.
It does raise the issue of what audio to output for a 10 inch hologram of a band - life size sound or small sound?
Speaking of holograms - even the visual is reproduced here: Is that not what we are trying to do, an audio hologram - audio gram?
YouTube
Audacity - and you do mean Real - Time DSP, right?
After searching for some DSP software including in-browser DSP and Windows programs, I finally settled on Audacity which I had aleady installed on my Ubuntu.
I loaded the track, and armed with some basic headphones, I did some equalization and compressing.
You recording engineers out there have really got your work cut out, and it is so fascinating really, I have had a real blast playing around with these basic controls.
First, I equalized, using the Bass Boost setting, then the RIAA setting, then I boosted the volume and seeing a lot of red, added compression.
The track is track 1 from Kenny Rogers "Eyes That See in the Dark" album from 1983 written and produced by the Bee Gees.
See from the screenshot the the red on this track appears at the bass and drum beats, so maybe some bass cut will help the thing play loud and also not distort.
Made 4 tracks and will try it next on my low power system to see if it improves the sound. The source is from a tape recorded to a CD, and used in wav format in Audacity.
All this is obviously not Real Time.
After searching for some DSP software including in-browser DSP and Windows programs, I finally settled on Audacity which I had aleady installed on my Ubuntu.
I loaded the track, and armed with some basic headphones, I did some equalization and compressing.
You recording engineers out there have really got your work cut out, and it is so fascinating really, I have had a real blast playing around with these basic controls.
First, I equalized, using the Bass Boost setting, then the RIAA setting, then I boosted the volume and seeing a lot of red, added compression.
The track is track 1 from Kenny Rogers "Eyes That See in the Dark" album from 1983 written and produced by the Bee Gees.
See from the screenshot the the red on this track appears at the bass and drum beats, so maybe some bass cut will help the thing play loud and also not distort.
Made 4 tracks and will try it next on my low power system to see if it improves the sound. The source is from a tape recorded to a CD, and used in wav format in Audacity.
All this is obviously not Real Time.
Attachments
The level of bass in a music signal needs to be in proportion to the size of the instruments being used.
Bass notes have a longer wavelength.
So if the band is small the bass will suffer.
This is also true of speaker wires. If they arent thick enough then they cannot carry the full spectrum of notes due to the cross section not being wide enough to allow the different frequencies to pass through.
Its called the skin effect in physics and means tgat the higher frequencies travel along the outer edge of the wire.
If its not thick enough then the higher frequencies get theortled by the BASS notes.
Mitch
Went through the preview of your book. The brief preview answered many of my questions and is really fascinating stuff. Much to read and many links to follow, however the link to Gedde's paper is
http://www.gedlee.com/Papers/directivity.pdf
and not Microsoft Azure Web App - Error 404
as listed in your Kindle edition book.
Also I see your note on listening levels in the book.
Thank you