Feature request: Fixed Maximum Hardware Gain (Preamp) Threshold

[Jean-Paul Petrosyan]

I have a suggestion concerning the Hardware Volume Control feature.

The thing is, I've taken to make use of Hardware Volume Control feature to adjust playback volume with precision not attainable otherwise, via Peamp slider (or manually entering numerical value in the box when tapping on it).
Don't know if it's a reasonable thing to do in principle, but I've definitely had some issues with this method in practice.

I've had a few accidents when due to some kind of a bug (one or two times) and my own mistakes and lack of attentiveness (while being sleepy/tired/drunk), when the hardware gain was raised dramatically and instantly while adjusting it in the process of music playback, which may have caused damage to my hardware (HP's drivers) and, more importantly, my hearing - due to acoustic shock/trauma. Obviously, It's me who is at fault here, so I have no right, nor any intention to blame the developer for this. But, I think it would benefit both end users as well as developer to make this software as foolproof as reasonably possible, at least regarding such features as volume controls, which have a potential to cause real harm to users if they're not careful enough. Of course, there will always be a possibility of raising volume to unsafe or even unreasonable levels and inflicting damage upon both your hardware and yourself with virtually any audio player. But what I'm talking about here is accidents.

Maybe it would be possible to add a feature which will let you set your own fixed threshold for maximum hardware gain value (like 0.0 or -10, for example) in the Audio Hardware Settings (for example), so that it will be impossible to surpass it accidentally while regulating it during playback (on the pic related screen), without changing the "Max value/Threshold" setting intentionally, which would require you to go to the dedicated Audio Hardware Settings menu.

[WitzyZed]

How is this different from the AGP (Automatic Gain Protection) ?

[blaubär]

The AGP limits the output level of the DSP by reducing the preamp, so that 0 dB isn't exceeded after the equalizer, crossfeed etc. See how the agp works

[Jean-Paul Petrosyan]

The AGP limits the output level of the DSP by reducing the preamp, so that 0 dB isn't exceeded after the equalizer, crossfeed etc. See how the agp works

How is this different from the AGP (Automatic Gain Protection) ?

To my understanding, AGP triggers when it detects overloaded signal, which exceeds 0 dB level, which happens when there's EQ enabled and a boost in certain frequency bands is present. If there's no such thing, it should not work, should it? AGP prevents distortion and clipping from pushing your hardware too hard, am I getting it right? But there's still enough room for signal to get very, very loud without distorting and triggering AGP. I'm talking more about the ability to manually set a threshold for maximum preamp/gain level (e.g. -15 dB), so that you cannot accidentally exceed it while adjusting volume via Preamp slider during listening, by accidentally sliding it too high or making a typo while entering it manually (e.g. I want to set it to - 19.1 dB, but accidentally set it to -11 dB or -1 dB, whereas - 20 dB is loud enough already, cause gain slider is maxed (it feels to me that keeping Preamp at default - 0.0 dB, and adjusting the Gain slider affects audio quality, while keeping Gain at maximum instead and adjusting volume with Preamp slider, while keeping it below 0.0 dB point - within negative values only, does not or does it to a lesser extent)) without accessing Settings and adjusting the maximum threshold there.

I'm starting to feel that my user-case and, thus, my request is pretty damn specific (or should I say "special", if you know what I mean), to put it mildly. In fact, might be way too specific to pay attention to it at all, I guess

[blaubär]

You could disable Settings > Audio Hardware > Hardware Gain and the Preamp switch, then yet another volume slider will appear ( on the eq screen instead of hardware gain or preamp ), which controls the volume after the DSP, and take that for fine control of volume, instead of the preamp or the hardware gain.

20200810_213428.jpg (9.36 KiB) Viewed 2718 times

[Jean-Paul Petrosyan]

You could disable Settings > Audio Hardware > Hardware Gain and the Preamp switch, then yet another volume slider will appear ( on the eq screen instead of hardware gain or preamp ), which controls the volume after the DSP, and take that for fine control of volume, instead of the preamp or the hardware gain.


20200810_213428.jpg

I know, but thank you for advice. I was under impression that this scenario results in quality loss, so that's why I resorted to such an unusual method. But, now that I have though about it, should a considerable decrease in Preamp level affect audio quality negatively? Could it be that I'm losing parts of a signal by setting hardware gain on maximum and lowering the Preamp, let's say, to - 25 dB? Some parts of music with high dynamic range may be lost anyways ( simply due too being to quiet to be picked up by auditory system ) when listening at low volume, obviously. But if something gets lost at the digital processing level, before the analog stage - shouldn't it be much more undesirable (and could it be happening with my scenario)?

Thank's to all for the answers already provided, and thank you in advance for your help.

[blaubär]

I was under impression that this scenario results in quality loss, so that's why I resorted to such an unusual method. But, now that I have though about it, should a considerable decrease in Preamp level affect audio quality negatively? Could it be that I'm losing parts of a signal by setting hardware gain on maximum and lowering the Preamp, let's say, to - 25 dB? Some parts of music with high dynamic range may be lost anyways ( simply due too being to quiet to be picked up by auditory system ) when listening at low volume, obviously. But if something gets lost at the digital processing level, before the analog stage - shouldn't it be much more undesirable (and could it be happening with my scenario)?

With 16 bit you've got a dynamic range of 96 db, but Neutron uses 32 bit or 64 bit, so you've got an dynamic range for calculations in the DSP of twice or four times that. So 25 db won't make any practical difference inside the DSP. And the digital data are transferred to the DAC with 32 bit ( at least on those platforms that support it, my Samsung S9 with Android 10 does it ). So 25 db from a total 192 db shouldn't make a difference either. I wouldn't expect anybody to be able to hear a difference between decreasing the volume before the DSP with the preamp or after the DSP with the gain, when in doubt use 64 bit.

[Jean-Paul Petrosyan]

I was under impression that this scenario results in quality loss, so that's why I resorted to such an unusual method. But, now that I have though about it, should a considerable decrease in Preamp level affect audio quality negatively? Could it be that I'm losing parts of a signal by setting hardware gain on maximum and lowering the Preamp, let's say, to - 25 dB? Some parts of music with high dynamic range may be lost anyways ( simply due too being to quiet to be picked up by auditory system ) when listening at low volume, obviously. But if something gets lost at the digital processing level, before the analog stage - shouldn't it be much more undesirable (and could it be happening with my scenario)?

With 16 bit you've got a dynamic range of 96 db, but Neutron uses 32 bit or 64 bit, so you've got an dynamic range for calculations in the DSP of twice or four times that. So 25 db won't make any practical difference inside the DSP. And the digital data are transferred to the DAC with 32 bit ( at least on those platforms that support it, my Samsung S9 with Android 10 does it ). So 25 db from a total 192 db shouldn't make a difference either. I wouldn't expect anybody to be able to hear a difference between decreasing the volume before the DSP with the preamp or after the DSP with the gain, when in doubt use 64 bit.

I have 64-Bit processing enabled just in case, although, if I get it right, it should not be of any benefit if the DSP options are not used. Yes, I know that theoretically there shouldn't be any (discernible) difference between these two options, but somehow I feel it and do so pretty clearly. It's funny, cause for me this is the case with some other things which should make no perceptible difference also (e.g. oversampling without any DSP enabled - sound becomes less "flat", the subjective soundstage increases in depth, while overall clarity decreases - this gets more prominent the higher the oversampling ratio is). My guess is that this may be due to the limitations of my phone's hardware as well as it's software "quirks" related to sound processing and it's interaction with Neutron, something of that nature.

Considering Preamp/Gain adjustments in their impact on sound - when I regulate volume via Preamp while keeping Hardware Gain on Max, I'm getting significantly more clarity in sound and overall technical performance of my headphones feels way better. Although the volume increase is percepted in a strange manner - low/mid/high frequencies seem to get louder more "unevenly" compared to the conventional way of volume adjustment (via Gain/Hardware Gain). I'm not able to volume match with enough precision, but at similar volume sound with Preamp is "thinner", while with Gain it is "fuller" although "muddier". My wild guess is that this is due to my phones inability to drive the headphones properly (36 Ohm impedance | 101 dB/mW sensitivity - they might not be power hungry, but are nevertheless demanding to the source to a considerable extent), I'm either being able to find the sweet spot, at the volume level that's barely enough for me where my dac/amp manages to perform OK with them, easier via Preamp adjustments, or I'm losing some part of the signal (from the low frequencies, maybe) due to Preamp manipulations and there's simply less for both my phone's dac/amp and headphones to tackle, so they perform better from the purely technical perspective with lesser load on them.

Here's my configuration just for reference:

[blaubär]

Perhaps there is some "sound optimization" built into the hardware / os and still active, some sort of "loudness" switch that enhances high and low frequencies dependent on the volume. or something of that sort. Perhaps you can take a look in the os-settions if somethine exists and how to disable it.