Maximum 50 dB gain in ALSA softvol plugin

Maximum 50 dB gain in ALSA softvol plugin

[Ricard Wanderlof]

Hi,

The ALSA softvol plugin has a maximum gain of 50 dB (MAX_DB_UPPER_LIMIT is 
fixed to 50 in pcm/pcm_softvol.c). Is there some technical reason for this 
limit? These days when >16 bit audio is becoming common, it's not 
unreasonable to have large gain figures, and that is in fact a usecase I 
have right now.

I'm looking through the CONVERT_AREA and MULTI_DIV_foo macros to see if 
there's something there that degenerates when the gain factor becomes too 
large, but I'd thought I'd ask too rather than just do reverse 
engineering.

/Ricard
-- 
Ricard Wolf Wanderlöf                           ricardw(at)axis.com
Axis Communications AB, Lund, Sweden            www.axis.com
Phone +46 46 272 2016                           Fax +46 46 13 61 30

Re: Maximum 50 dB gain in ALSA softvol plugin

[Takashi Iwai]

On Fri, 12 Jan 2018 17:02:03 +0100,
Ricard Wanderlof wrote:
> 
> 
> Hi,
> 
> The ALSA softvol plugin has a maximum gain of 50 dB (MAX_DB_UPPER_LIMIT is 
> fixed to 50 in pcm/pcm_softvol.c). Is there some technical reason for this 
> limit? These days when >16 bit audio is becoming common, it's not 
> unreasonable to have large gain figures, and that is in fact a usecase I 
> have right now.
> 
> I'm looking through the CONVERT_AREA and MULTI_DIV_foo macros to see if 
> there's something there that degenerates when the gain factor becomes too 
> large, but I'd thought I'd ask too rather than just do reverse 
> engineering.

There is no more reason than "some reasonable limit".


Takashi

Re: Maximum 50 dB gain in ALSA softvol plugin

[Ricard Wanderlof]

On Fri, 12 Jan 2018, Takashi Iwai wrote:

> > The ALSA softvol plugin has a maximum gain of 50 dB (MAX_DB_UPPER_LIMIT is 
> > fixed to 50 in pcm/pcm_softvol.c). Is there some technical reason for this 
> > limit? These days when >16 bit audio is becoming common, it's not 
> > unreasonable to have large gain figures, and that is in fact a usecase I 
> > have right now.
> > 
> > I'm looking through the CONVERT_AREA and MULTI_DIV_foo macros to see if 
> > there's something there that degenerates when the gain factor becomes too 
> > large, but I'd thought I'd ask too rather than just do reverse 
> > engineering.
> 
> There is no more reason than "some reasonable limit".

Ok, my suggestion would be set MAX_DB_UPPER_LIMIT to 90 dB. 90 dB comes 
out as 31622, which still is manageable when shifted up by VOL_SCALE_SHIFT 
(= 16), fitting in a 32-bit int. Thus it works in MULTI_DIV_short() too, 
where it is multiplied with the 16-bit sample value yielding an int.

I'll try it out and prepare a patch.

/Ricard
-- 
Ricard Wolf Wanderlöf                           ricardw(at)axis.com
Axis Communications AB, Lund, Sweden            www.axis.com
Phone +46 46 272 2016                           Fax +46 46 13 61 30

Re: Maximum 50 dB gain in ALSA softvol plugin

[Takashi Iwai]

On Mon, 15 Jan 2018 14:14:13 +0100,
Ricard Wanderlof wrote:
> 
> 
> On Fri, 12 Jan 2018, Takashi Iwai wrote:
> 
> > > The ALSA softvol plugin has a maximum gain of 50 dB (MAX_DB_UPPER_LIMIT is 
> > > fixed to 50 in pcm/pcm_softvol.c). Is there some technical reason for this 
> > > limit? These days when >16 bit audio is becoming common, it's not 
> > > unreasonable to have large gain figures, and that is in fact a usecase I 
> > > have right now.
> > > 
> > > I'm looking through the CONVERT_AREA and MULTI_DIV_foo macros to see if 
> > > there's something there that degenerates when the gain factor becomes too 
> > > large, but I'd thought I'd ask too rather than just do reverse 
> > > engineering.
> > 
> > There is no more reason than "some reasonable limit".
> 
> Ok, my suggestion would be set MAX_DB_UPPER_LIMIT to 90 dB. 90 dB comes 
> out as 31622, which still is manageable when shifted up by VOL_SCALE_SHIFT 
> (= 16), fitting in a 32-bit int. Thus it works in MULTI_DIV_short() too, 
> where it is multiplied with the 16-bit sample value yielding an int.
> 
> I'll try it out and prepare a patch.

Well, but +90dB gain doesn't look like a normal usage pattern.
Why do you have to use softvol plugin for achieving that specific
purpose?


Takashi

Re: Maximum 50 dB gain in ALSA softvol plugin

[Ricard Wanderlof]

On Mon, 15 Jan 2018, Takashi Iwai wrote:

> > Ok, my suggestion would be set MAX_DB_UPPER_LIMIT to 90 dB. 90 dB comes 
> > out as 31622, which still is manageable when shifted up by VOL_SCALE_SHIFT 
> > (= 16), fitting in a 32-bit int. Thus it works in MULTI_DIV_short() too, 
> > where it is multiplied with the 16-bit sample value yielding an int.
> > 
> > I'll try it out and prepare a patch.
> 
> Well, but +90dB gain doesn't look like a normal usage pattern. Why do 
> you have to use softvol plugin for achieving that specific purpose?

The usecase we have actually calls for 55 dB of gain maximum, but I was 
thinking that looking at the code the maths can actually handle 90 dB so 
it would be good 'reasonable limit' - it's a nice round number and fits in 
16 bits (signed).

The usecase we have is using digital MEMS microphones ((2S connected, 
with no integral amplification) and when the sound source is far away, a 
fair amount of gain can be needed.

/Ricard
-- 
Ricard Wolf Wanderlöf                           ricardw(at)axis.com
Axis Communications AB, Lund, Sweden            www.axis.com
Phone +46 46 272 2016                           Fax +46 46 13 61 30

Re: Maximum 50 dB gain in ALSA softvol plugin

[Takashi Iwai]

On Mon, 15 Jan 2018 14:49:38 +0100,
Ricard Wanderlof wrote:
> 
> 
> On Mon, 15 Jan 2018, Takashi Iwai wrote:
> 
> > > Ok, my suggestion would be set MAX_DB_UPPER_LIMIT to 90 dB. 90 dB comes 
> > > out as 31622, which still is manageable when shifted up by VOL_SCALE_SHIFT 
> > > (= 16), fitting in a 32-bit int. Thus it works in MULTI_DIV_short() too, 
> > > where it is multiplied with the 16-bit sample value yielding an int.
> > > 
> > > I'll try it out and prepare a patch.
> > 
> > Well, but +90dB gain doesn't look like a normal usage pattern. Why do 
> > you have to use softvol plugin for achieving that specific purpose?
> 
> The usecase we have actually calls for 55 dB of gain maximum, but I was 
> thinking that looking at the code the maths can actually handle 90 dB so 
> it would be good 'reasonable limit' - it's a nice round number and fits in 
> 16 bits (signed).
> 
> The usecase we have is using digital MEMS microphones ((2S connected, 
> with no integral amplification) and when the sound source is far away, a 
> fair amount of gain can be needed.

But why do you need 90dB?  "Because it can" is no good reason to
increase such a safe limit.


Takashi

Re: Maximum 50 dB gain in ALSA softvol plugin

[Ricard Wanderlof]

On Mon, 15 Jan 2018, Takashi Iwai wrote:

> > The usecase we have actually calls for 55 dB of gain maximum, but I was 
> > thinking that looking at the code the maths can actually handle 90 dB so 
> > it would be good 'reasonable limit' - it's a nice round number and fits in 
> > 16 bits (signed).
> > 
> > The usecase we have is using digital MEMS microphones ((2S connected, 
> > with no integral amplification) and when the sound source is far away, a 
> > fair amount of gain can be needed.
> 
> But why do you need 90dB?  "Because it can" is no good reason to
> increase such a safe limit.

No, I don't need 90 dB. The specific usecase I have calls for 55 dB, so I 
could of course just change the maximum from 50 to 55 dB. Or perhaps with 
a bit of margin, such as 60 dB or 70 dB.

They way I see it though, softvol is just an amplifier component. It's up 
to the user to use it for whatever purpose they want. And there is no 
inharent danger in allowing large gains, as the signal will just clip at 
full scale, it's not like having a power amplifier with infinite power 
resources. So the upper limit should be bound by something technical, like 
an internal processing stage overflowing, which is why I suggested 90 dB, 
as it is at slightly more than 90 dB that a 16 bit signed short overflows, 
16-bit shorts being part of the current set of calculations in the 
MULTI_DIV_ macros. So 90 dB seemed to me a limit that is related to the 
technical limitations of the algorithm employed.

Note that I'm not suggesting changing the default gain range which I 
believe is -50 to 0 dB. In order to utilize the large gain, someone must 
purposefully enter a 'max_dB' value in /etc/asound.conf and then set the 
gain to that value, so there would seem little risk of setting a large 
gain by mistake.

/Ricard
-- 
Ricard Wolf Wanderlöf                           ricardw(at)axis.com
Axis Communications AB, Lund, Sweden            www.axis.com
Phone +46 46 272 2016                           Fax +46 46 13 61 30

Re: Maximum 50 dB gain in ALSA softvol plugin

[Jaroslav Kysela]

Dne 15.1.2018 v 14:49 Ricard Wanderlof napsal(a):
> 
> On Mon, 15 Jan 2018, Takashi Iwai wrote:
> 
>>> Ok, my suggestion would be set MAX_DB_UPPER_LIMIT to 90 dB. 90 dB comes 
>>> out as 31622, which still is manageable when shifted up by VOL_SCALE_SHIFT 
>>> (= 16), fitting in a 32-bit int. Thus it works in MULTI_DIV_short() too, 
>>> where it is multiplied with the 16-bit sample value yielding an int.
>>>
>>> I'll try it out and prepare a patch.
>>
>> Well, but +90dB gain doesn't look like a normal usage pattern. Why do 
>> you have to use softvol plugin for achieving that specific purpose?
> 
> The usecase we have actually calls for 55 dB of gain maximum, but I was 
> thinking that looking at the code the maths can actually handle 90 dB so 
> it would be good 'reasonable limit' - it's a nice round number and fits in 
> 16 bits (signed).
> 
> The usecase we have is using digital MEMS microphones ((2S connected, 
> with no integral amplification) and when the sound source is far away, a 
> fair amount of gain can be needed.

Wow, the signal quality from few bits per sample must be really good ;-)
It would be better to use ADC with higher resolution (24-bit) and low
noise floor (but I see the possible costs requirements). This is why we
have such good audio in collaborative conferences :-)

					Jaroslav

-- 
Jaroslav Kysela <perex@perex.cz>
Linux Sound Maintainer; ALSA Project; Red Hat, Inc.

Re: Maximum 50 dB gain in ALSA softvol plugin

[Takashi Iwai]

On Mon, 15 Jan 2018 15:32:13 +0100,
Ricard Wanderlof wrote:
> 
> 
> On Mon, 15 Jan 2018, Takashi Iwai wrote:
> 
> > > The usecase we have actually calls for 55 dB of gain maximum, but I was 
> > > thinking that looking at the code the maths can actually handle 90 dB so 
> > > it would be good 'reasonable limit' - it's a nice round number and fits in 
> > > 16 bits (signed).
> > > 
> > > The usecase we have is using digital MEMS microphones ((2S connected, 
> > > with no integral amplification) and when the sound source is far away, a 
> > > fair amount of gain can be needed.
> > 
> > But why do you need 90dB?  "Because it can" is no good reason to
> > increase such a safe limit.
> 
> No, I don't need 90 dB. The specific usecase I have calls for 55 dB, so I 
> could of course just change the maximum from 50 to 55 dB. Or perhaps with 
> a bit of margin, such as 60 dB or 70 dB.
> 
> They way I see it though, softvol is just an amplifier component. It's up 
> to the user to use it for whatever purpose they want. And there is no 
> inharent danger in allowing large gains, as the signal will just clip at 
> full scale, it's not like having a power amplifier with infinite power 
> resources. So the upper limit should be bound by something technical, like 
> an internal processing stage overflowing, which is why I suggested 90 dB, 
> as it is at slightly more than 90 dB that a 16 bit signed short overflows, 
> 16-bit shorts being part of the current set of calculations in the 
> MULTI_DIV_ macros. So 90 dB seemed to me a limit that is related to the 
> technical limitations of the algorithm employed.
> 
> Note that I'm not suggesting changing the default gain range which I 
> believe is -50 to 0 dB. In order to utilize the large gain, someone must 
> purposefully enter a 'max_dB' value in /etc/asound.conf and then set the 
> gain to that value, so there would seem little risk of setting a large 
> gain by mistake.

Sure, the default shouldn't be changed, otherwise we'll hit many
complaints :)

OTOH, the amplification in softvol is really dumb, and such a high
gain like 90dB is doubtful whether it's really useful.  As Jaroslav
already suggested, we need a better setup to get more meaningful
results.


Takashi

Re: Maximum 50 dB gain in ALSA softvol plugin

[Ricard Wanderlof]

On Mon, 15 Jan 2018, Jaroslav Kysela wrote:

> > The usecase we have actually calls for 55 dB of gain maximum, but I was 
> > thinking that looking at the code the maths can actually handle 90 dB so 
> > it would be good 'reasonable limit' - it's a nice round number and fits in 
> > 16 bits (signed).
> > 
> > The usecase we have is using digital MEMS microphones ((2S connected, 
> > with no integral amplification) and when the sound source is far away, a 
> > fair amount of gain can be needed.
> 
> Wow, the signal quality from few bits per sample must be really good ;-)
> It would be better to use ADC with higher resolution (24-bit) and low
> noise floor (but I see the possible costs requirements). This is why we
> have such good audio in collaborative conferences :-)

Ok, I wasn't expecting this kind of discussion on this subject ... 

It's true that for best performance, analog microphones (even analog MEMS) 
trump digital MEMS microphones when it comes to SNR, but in many cases, 
the SNR of the room itself will be the limiting factor anyway.

The point is this: a typical digital MEMS microphone has 0 dBFS output at 
120 dB SPL. Looking at one typical recording usecase, the background noise 
level in a quiet room is 40 or 50 dB SPL, with speech at a distance of a 
meter or so being around 60 dB SPL. So in this case we're 60 dB below full 
scale for the speech signal, with the acoustic noise floor being 10 or 20 
dB below that. The noise floor of a MEMS microphone of this type is 
equivalent to about 30 dB SPL, so the acoustic noise is still higher than 
the electrical noise in the system.

Now, assuming we actually want to hear what is going on, so we bring up 
the gain by 60 dB in order to get the speech to full scale. That 
means that the noise floor comes up too, and there's not much we can do 
about that anyway as it's part of the acoustic scene we're capturing. The 
fact that we loose bit depth is not of much consequence, as the acoustic 
noise in the scene is above the resulting quantization noise anyway, in 
much the same way that adding dither to a signal masks the quantization 
noise. The acoustic SNR is 10-20 dB so we could actually represent the 
resulting signal with 8 bits and still be fine.

A side note here is that many PC's, especially laptops, do not provide 
much playback gain, so there is a point in bringing up signals to close to 
0 dBFS.

The bottom line is that, yes, we loose bit depth when we apply gain, but 
that itself doesn't impact the sound quality, as long as the noise level 
is above the quantization noise to start with, which it is in many 
microphone cases. And we really might need a large gain in certain 
situations.

On Mon, 15 Jan 2018, Takashi Iwai wrote:

> [ ... ] the amplification in softvol is really dumb, and such a high 
> gain like 90dB is doubtful whether it's really useful.  As Jaroslav 
> already suggested, we need a better setup to get more meaningful 
> results.

In what way is it dumb? Amplification is just a multiplication with a gain 
factor, and the softvol plugin seems to do that fine. I just tested 
patching it to 90 dB gain max, and applying a (24 bit) sine wave at -90 dB 
and amplifying it by 90 dB using softvol, and I couldn't see any odd 
artfacts.

Of course, for the usecase I described, an AGC of some form, or dynamic 
compressor would probably be better, but for a static gain, softvol could 
very well be employed.

Ok ... bottom line ... I'd like to increase the maximum potential gain for 
softvol above the current 50 dB. I suggested 90 dB because that's how much 
the algorithm can handle, and that's what I figured the MAX_DB_UPPER_LIMIT 
in pcm/pcm_softvol.c should reflect. Furthermore, I can't see any problem 
with increasing the limit; it does not degenerate the algorithm or cause 
potential problems for existing users. Admittedly, this is not kernel 
code, but if it were I would expect that the point of view would be that 
the code should not make assumptions on user's policies but only its own 
technical limitations. And if the input is 32 bit audio, we'd still end up 
with about 16 bits of resolution when applying 90 dB of gain, so 
mathematically it's not unreasonable.

The only problem I can foresee is that it means that all future changes to 
the algorithm might need to accommodate the specified maximum gain to 
avoid annoying users who are actually using it, which might be a problem 
somewhere down the line.

But for my usecase I'd be happy just pushing the limit to 60 or 70 dB if 
that's more acceptable.

/Ricard
-- 
Ricard Wolf Wanderlöf                           ricardw(at)axis.com
Axis Communications AB, Lund, Sweden            www.axis.com
Phone +46 46 272 2016                           Fax +46 46 13 61 30

Re: Maximum 50 dB gain in ALSA softvol plugin

[Ricard Wanderlof]

On Mon, 15 Jan 2018, Takashi Iwai wrote:

> OTOH, the amplification in softvol is really dumb, and such a high
> gain like 90dB is doubtful whether it's really useful.  As Jaroslav
> already suggested, we need a better setup to get more meaningful
> results.

Hi Takashi,

I didn't really get a definitive response to my elaboration last week 
regarding the maximum allowable gain value for the softvol plugin.

First of all, regarding the algorithm itself used, as it stands today it 
can handle a maximum of 90.3 dB gain without falling apart (as the gain 
factor is held in a signed integer). Changing the maximum allowable gain 
from 50 dB to 90 dB does not impact the algorithm or existing 
functionality in any way, it just changes the allowable maximum to 
correspond to the technical limit of the current algorithm. The limit is 
not visible to the user anyway until he tries to configure a max_dB that 
is above 50 (and then only in the shape of an error message). So from a 
purely technical viewpoint I cannot understand any opposition.

You mentioned that the amplification is really dumb, I'm not sure what the 
problem is, perhaps you could elaborate, essentially a gain is a 
multiplication with a gain factor, and that's what softvol does.

Another issue is if such a large gain is useful. I would say that is up to 
the user, and of course, if you have a weak signal to start with, with, 
say, only 8 bits of resolution, adding 90 dB of gain is going to mean 
there's a lot of noise bits there, but the signal quality is of course not 
worse than it was to start with, just louder, and that's how any signal 
(digital or analog) behaves when gain is added.

But the bottom line of course is that you are the maintainer. So what 
would you consider a reasonable maximum gain that's more than 50 dB. 70 
dB?

/Ricard
-- 
Ricard Wolf Wanderlöf                           ricardw(at)axis.com
Axis Communications AB, Lund, Sweden            www.axis.com
Phone +46 46 272 2016                           Fax +46 46 13 61 30