Optimal relationship between device volume and amplifier volume
Mar 21, 2019 at 8:46 PM Thread Starter Post #1 of 7

superdragon

Head-Fier
Joined
Apr 4, 2014
Posts
75
Likes
2
I'm streaming to a Bluetooth receiver which gets to a small digital amp which feeds a loudspeaker.

To get the best quality sound, is it recommended to have a high device volume and low amp volume, or the reverse, or something in between? Is there a known formula; something like 70% device volume, them turn up the amp until the volume suits you? Does it not make a difference at all? Do some different mixtures produce the same volume and same quality?
 
Mar 22, 2019 at 12:26 AM Post #2 of 7
I'm streaming to a Bluetooth receiver which gets to a small digital amp which feeds a loudspeaker.

To get the best quality sound, is it recommended to have a high device volume and low amp volume, or the reverse, or something in between? Is there a known formula; something like 70% device volume, them turn up the amp until the volume suits you? Does it not make a difference at all? Do some different mixtures produce the same volume and same quality?

I've thought about that too. If I had to hazard a guess, I'd say the volume adjuster on the Bluetooth radio is an attenuator that probably works like most digital volume controls, by attenuating bits. For best sound, if your Bluetooth radio is working in 16 bits it's probably best to keep the Bluetooth radio's volume at 100% and control the volume with the amp. OTOH if your Bluetooth radio is working in 24 bits, it may not matter. <--- just a guess
 
Mar 22, 2019 at 1:21 AM Post #3 of 7
For digital volume controls, all the way up as upstateguy said, or at least until you have 14 or so bits. For amps, somewhere around 12-3 o'clock is usually ideal. Too low and inaccuracies in the volume pot become apparent and too high and you might drive it into clipping. You'll also want to stop short of any audible noise floor as well.
 
Last edited:
Mar 23, 2019 at 12:08 PM Post #4 of 7
Does it make any difference if for instance you are running from a source like an Iphone or a DAP into a Bluetooth adapter like a BTR1K? What I usually do is max out the volume from the source and control the volume from the BTR1k...
 
Mar 23, 2019 at 8:20 PM Post #5 of 7
whatever sounds fine. and if everything seems to sound fine, then great. the general rule of thumb is still to try and leave the digital gain alone, or at least use it within reason. if you end up with a setup that requires to attenuate the digital volume by 80dB, well something is really wrong in your setup.
but beyond that, there are just so many potential unknowns. depending on the codec used, how far the signal is from 0dB won't matter much. but what may matter is how much total dynamic range is left for the music anytime we're not in a 32bit float or 64bit environment. if we're limited to 16bit at some point in the playback chain and we digitally attenuate a bunch, the least significant bits could be lost in the process. but if the all thing is worked out at 24 or 32bit, we're likely to have ample headroom anyway. and maybe the BT codec is lossy and starts to discard stuff as soon as -60dB anyway so who cares about least significant bits^_^. or maybe the BT receiver only has digital volume control itself and it all ends up being the same. or not because it's in a 32/64bit scheme so in almost all realistic conditions, the music will be fine.
and that's not even considering how most BT devices have to be small, light and low consumption so their amp section is often crap with a high level of noise that will more than cover quantization errors.

unless you know exactly what does what, I would suggest to "forgetaboutit". just use common sense, like avoiding to reduce something by 60dB and expect the amp to be able to compensate all of it for absolutely no reason. but beyond that, just set the general volume at the amp if there is such a choice and fine tune the listening level however is more convenient for you.
 
Mar 24, 2019 at 1:53 AM Post #6 of 7
Does it make any difference if for instance you are running from a source like an Iphone or a DAP into a Bluetooth adapter like a BTR1K? What I usually do is max out the volume from the source and control the volume from the BTR1k...

whatever sounds fine. and if everything seems to sound fine, then great. the general rule of thumb is still to try and leave the digital gain alone, or at least use it within reason. if you end up with a setup that requires to attenuate the digital volume by 80dB, well something is really wrong in your setup.
but beyond that, there are just so many potential unknowns. depending on the codec used, how far the signal is from 0dB won't matter much. but what may matter is how much total dynamic range is left for the music anytime we're not in a 32bit float or 64bit environment. if we're limited to 16bit at some point in the playback chain and we digitally attenuate a bunch, the least significant bits could be lost in the process. but if the all thing is worked out at 24 or 32bit, we're likely to have ample headroom anyway. and maybe the BT codec is lossy and starts to discard stuff as soon as -60dB anyway so who cares about least significant bits^_^. or maybe the BT receiver only has digital volume control itself and it all ends up being the same. or not because it's in a 32/64bit scheme so in almost all realistic conditions, the music will be fine.
and that's not even considering how most BT devices have to be small, light and low consumption so their amp section is often crap with a high level of noise that will more than cover quantization errors.

unless you know exactly what does what, I would suggest to "forgetaboutit". just use common sense, like avoiding to reduce something by 60dB and expect the amp to be able to compensate all of it for absolutely no reason. but beyond that, just set the general volume at the amp if there is such a choice and fine tune the listening level however is more convenient for you.

I mostly stream from my Microsoft Surface Pro 4 to a Bluetooth receiver (https://www.amazon.com/gp/product/B01EHSX28M/) and then to a small digital amp and to a speaker. With this configuration, the Windows volume output at 100% isn't the best. This only allows me to turn the amp volume up a drop and the sound feels a little too powerful. I feel comfortable with system volume somewhere between 68 - 72% and then adjusting the amp volume accordingly. Within that range and lower, there are different mixes of system/amp settings that appear to produce the same results. I was curious as to the sound science (better or worse) with the different mixes, but from he replies, I'm comfortable that there are no differences. It is though fully clear and obvious that high amp volume and low system volume don't produce good results.

Anyway, this topic has always interested me - even back in the "old days" with line in connections to powered speakers. I'm assuming the same logic applies in those situations.

Ty.
 
Mar 24, 2019 at 3:17 AM Post #7 of 7
well if we disregard Bluetooth or what it might do to the signal, and only look at some ideal concept of digital vs analog volume setting, then it's almost always better to use analog volume from a fidelity standpoint. but that's only the pretty theory when no other problem can occur.
in practice, you mention the obvious issue of sometimes having too much gain on the amp, so handling the volume knob is no fun at all. there is also a sort of rule of thumb about trying to avoid the first 1/3rd of the knob's course because on analog volume, some channel imbalance is more likely to increase there(that's really a case by case matter and only by measuring can we know for a fact). when playing a signal at 100%, there is also the possibility to clip the signal a little(intersample clipping), or simply because we clip it ourselves with some EQ boost or other digital processing. and obviously it would be a little silly to end up clipping the loudest sounds on the track just from fear that altering the least significant bits might be audible.

so real life situations may require to drop the ideal maxed out digital and volume setting only from an analog control. depends on the circumstances. luckily, we usually need to make a pretty mess for it to clearly affect our listening experience. I mention clipping, which should be the very worst thing to happen to an audio signal. well, so long as it's rather small and mostly on the low frequencies, people may not notice. there is even a psychoacoustic process where we reconstruct the sine from expectation in our mind and feel like the signal is louder and not clipped. so I reiterate, don't worry too much about that volume level thing. and if you do care anyway, maybe get a mic and try to measure what comes out of your speakers. it's the sure way to know once and for all. just that it's also sort of a PITA ^_^.
 

Users who are viewing this thread

Back
Top