Superampman
AK Member
I have no problem noticing a 1db increase in volume. So where does the 3db rule come from? I'm currently using a Meridian 501 pre with 1db increments. 3db sounds very significant. Am I alone here?
-
Please note that there are a few updates and clarifications made in the Audiokarma Rules, mostly relating to advertising and the addition of the new "Paying it Forward" & "Giving back" forums in the AudioKarma Audio Marketplace section.
1db is clearly perceptible
- Thread starter Superampman
- Start date
Register to hide this ad
Dennis Gardner
Jazz Nut
You had better be able to hear the difference, or I would have ask for the mods to take back your screen name.......
You should be able to hear 1 db increase or decrease of a steady single tone. With music, such a difference may or may not be that noticeable, it really depends on the kind of music and the volume level one is starting from.
What I have found surprisingly easy to hear is a quite small difference in volume between two channels (channel balance). When I had a Mark Levinson No. 32 preamp in my system, I could easily hear a .2 db change in level in one channel (perceived as a movement of a centered vocalist in a stereo recording or movement of the entire signal with monophonic sources). At first I thought it made no sense that the Levinson permitted such small incremental changes in volume, but, since the same set of resistors is used to provide BOTH volume and channel balance change (the signal in each channel goes through one set of resistors, not one for volume and one for balance), the small incremental steps are needed for purposes of providing the necessary degree of adjustment for channel balance.
What I have found surprisingly easy to hear is a quite small difference in volume between two channels (channel balance). When I had a Mark Levinson No. 32 preamp in my system, I could easily hear a .2 db change in level in one channel (perceived as a movement of a centered vocalist in a stereo recording or movement of the entire signal with monophonic sources). At first I thought it made no sense that the Levinson permitted such small incremental changes in volume, but, since the same set of resistors is used to provide BOTH volume and channel balance change (the signal in each channel goes through one set of resistors, not one for volume and one for balance), the small incremental steps are needed for purposes of providing the necessary degree of adjustment for channel balance.
Superampman
AK Member
The moniker is a take off from 'The Leach Superamp' mono of which I have a pair. Used to be my favorite. Still is one of them. But I have gathered over the years that 3db is the recognized standard difference required for the average individual to perceive a change in volume. I think it's wrong. Simple as that. I guess I'm looking for consensus.
Do an experiment.
Listen to something. Then have a friend change that something by -1, 0, or 1 dB. Without knowing if or what the change was, listen again. Do it a bunch of times and statistically determine if you can identify the change accurately. Then try it using a change of 3 dB. :scratch2:
Excellent suggestion from toddalin (which also alludes to whoaru's comment about it being easier to know there is a change when you are the one making it).
That said, 1db is generally defined as the smallest discernible change in perceived loudness. 3db is considered to be a material, but not dramatic, change...one which would be identified by most people as a "real change". In other words, if someone in the next room says "hey, turn up the music", they will recognize 3db as being "turned up". They would probably think that you failed to do so at 1dB (or that you were being a wise guy).
And, as you probably know, 10dB is considered a doubling in perceived loudness.
To summarize, I think you are starting out with a misconception; specifically, that 3dB is considered to be the smallest discernible change. Not the case.
That said, 1db is generally defined as the smallest discernible change in perceived loudness. 3db is considered to be a material, but not dramatic, change...one which would be identified by most people as a "real change". In other words, if someone in the next room says "hey, turn up the music", they will recognize 3db as being "turned up". They would probably think that you failed to do so at 1dB (or that you were being a wise guy).
And, as you probably know, 10dB is considered a doubling in perceived loudness.
To summarize, I think you are starting out with a misconception; specifically, that 3dB is considered to be the smallest discernible change. Not the case.
Last edited:
The other thing to consider is are those marks on the volume control a calibrated 1 db increment? Wild guess says they probably are not. Its also a question of 1db measured from where? 1db increase in signal input to the power amplifier should result in more than 1 db change in signal from the amplifier output since there is gain in the amp. Then there is speaker efficiency to factor in, etc. Really you'd need to set up some test gear and measure the SPL in the room, and then adjust by +1db to make this a fair test.
ggrey
Sonically Insane
It really does matter with what the context of the increase is in. I record music in a multi track format ( see avatar). If I increase one of the harmonies in a 3 part harmony mix by even a quarter db, it makes a big difference that anyone can clearly hear.
Although I am not sure I could distinguish a quarter db in volume level of an entire mix.
Although I am not sure I could distinguish a quarter db in volume level of an entire mix.
mech986
Lunatic Member
This. If you are doing blind testing and not knowing if or when changes are made, audible hearing memory can be tricky. If you are comparing steady state and trying to find the difference between two levels (say left and right), the ear is more discriminating for awhile. Listen for a longer time and the ear may tend to average things out.
A corollary test may be one where a level is arbitrarily raised some db above reference and then reset to another level, and the listener is asked to tell whether there is a difference in the current verses the original level. Same with decreasing then raising.
At that point, differences can be heard, but what level changes it take to become audible and memorable may be more than a db.
IN the case where harmony is the object, the subjective levels to match the harmony levels at different frequencies can be subjectively matched, but trying to do it strictly by meters and signal levels belies the differences in energy content at different levels and the ear's sensitivity to it or the overtones. That's why music is sometimes so tricky to balance.
HTHMAN
Super Member
If you really want to know what you can hear, use an SPL meter at your listening position. Have one friend read the meter and another change the volume. You should not be able to view either. Play a tone or white noise as music varies too much for a true reading. Note when you can tell the difference in sound level and have the friend note the change indicated on the meter.
Superampman
AK Member
Misconceptions abound here on AK. As long as they're corrected, wasted energy is a non issue.
Thank you all for your effort.
ra7c7er
Hoarding records since 1990
Exactly power of perception. If you put someone in a room and slowly turned up the volume, heat, ac (down), etc. 95+% of percent of people won't notice till you go up or down a couple decibels or degrees. But if you put them in the same room and have them see a gauge that shows temps or db they will claim to feel the difference even if their isn't actually one. This type of behavior has been studied a lot. Many colleges and universities us this simple type of study when teaching students how to run studies so it's done a lot.
fred soop
Super Member
I believe that much of this was determined many years ago with numerous test subjects by AT&T Labs. Fletcher-Munson curves were also determined by testing many subjects. Since everyone is different, the results were averaged but the definition of 1 dB was the minimum discernible change using pure tones. It would also have to depend on the frequency.
As for mixing, and changing the level of one channel, you now have a signal in the mix to compare with so the discernible change would be much smaller, but that is not the definition of 1 dB.
As for mixing, and changing the level of one channel, you now have a signal in the mix to compare with so the discernible change would be much smaller, but that is not the definition of 1 dB.
SuperjazzyJa
18 Wheel Member
Excellent points! The volume control on my receiver goes up and down in half dB increments and I can hear the difference regardless of what I am listening to. It would seem that what's coming out of the speakers would be changed by more than just half a decibel if the difference is audible.
Brian
An Old Geezer
The Bell test that lead to the various interpretations as to this issue was using steady tone tests and established that 3db was the average increase or decease in level at 1kz that an average person could perceive. The test was related to the development of the ear and mic pieces used in a new generation of phone; one where a person was not required to yell into.
It determined that a 1 db change could be discerned but, not by an average person other than using a steady tone and 3db was the practical minimum for voice as it is a complex waveform and not uniform in level.
This study lead to the beginnings of the study of signal compression to prevent overloading a system as well as new designs and specifications for amplifying telephone transmission signals.
Over the years the observation did seem to take on a life of its own and the word "average" dropped out of the application and discussions by engineers, manufacturers and laypersons. Its real relevance was that it was found to take 2x the average power to realize a 3db gain and is most useful today in helping those who are considering buying or updating whether the cost between 2 units is worth the premium for only a small increment of power increase.
It determined that a 1 db change could be discerned but, not by an average person other than using a steady tone and 3db was the practical minimum for voice as it is a complex waveform and not uniform in level.
This study lead to the beginnings of the study of signal compression to prevent overloading a system as well as new designs and specifications for amplifying telephone transmission signals.
Over the years the observation did seem to take on a life of its own and the word "average" dropped out of the application and discussions by engineers, manufacturers and laypersons. Its real relevance was that it was found to take 2x the average power to realize a 3db gain and is most useful today in helping those who are considering buying or updating whether the cost between 2 units is worth the premium for only a small increment of power increase.
Similar threads
