[CraftyClown]

Hi all,
 
Is it better to use a lower gain and higher volume, or a higher gain and lower volume?
 
I hear much talk about some headphones not sounding their best, even at high volumes unless they are sufficiently powered, however coming from a video background I am acutely aware that adding electronic gain to boost a signal raises the noise floor.
 
Is this a potential trade off? Or have I got the science confused?
 
Happy to be schooled

 

 

[castleofargh]

to get a proper response, I believe we would need to measure the output using a given set up, or at least know very well what design is used in the amp and have plenty of reliable specs.
often, changing the gain doesn't really show much difference unless you're borderline of clipping/distortions by reaching the limits for nominal behavior of some components. and if you've bought a fitting amp for a headphone and DAC, then you should have a lot of amplitude to fool around before reaching those types of situations.
 
on some amps the gain settings exist to match the input voltage of the DAC(that usually wanders between 1 and 3.5Vrms depending on DAC models). so that's a different matter but input gain is a thing.
 
the type of volume control could also play a role, some will have channel imbalance at lower volume setting, so reducing the gain lets you turn the knob higher and often get better balance for that kind of knobs.
about the added noise from higher gain, you're right on principle. just boosting the voltage more to then just force the volume knob to attenuate a lot more too, that doesn't seem to make much sense and could potentially increase all kinds of noises. but again that could depend on the design, the O2 amp is doing just that, keeping stuff high from the get go with the aim of keeping as much dynamic as possible all the way through. and even if it can seem counter intuitive, the result is a surprisingly low noise floor. but of course by keeping everything close to the max, if the DAC has a higher voltage than expected, then you go straight into horrible distortions and a matching input becomes more important than it will be on typical amp designs. so again, different designs can have different impacts and different needs.
 
 
and those are only a few ideas of what could matter on the top of my head, so it's hard to make an absolute claim on this subject.
 
still the general consensus for voltage amp, everything else considered stable and properly paired, is to use the lower gain and increase the volume control. then when you reach the maxed out loudness and still want it louder, you switch to a higher gain setting and find out if it's ok. that is the most common logic behind an output gain implementation.
 
the sufficiently powered part you talk about is both true and total BS at the same time. for typical voltage amp, the gain is a voltage gain and voltage is directly related to loudness. so at a given loudness in a proper system, the load(headphone's impedance) and loudness(voltage) are the limiting factors in the circuit and will determine the power applied for a given amp. it means that the louder you will need to go to enjoy your music with a sound system, the more power you will need. that much is intuitive and everybody gets it.
but it also means that for a specific loudness, the power is fixed into a given headphone. so if you can go loud enough on low gain, then high gain is not going to send more power into the headphone once the volume is set at the same loudness. and that part is usually misunderstood and leads to people talking nonsense about more power. often even mistaking the specs of maximum available power for the actual power used, and other funny stuff like that. a little like thinking that a 800W PSU inside a computer is going to send 800W to the DVD player. that's just nonsense.
 
of course voltage isn't everything and the amp could fail to deliver enough current, but will increasing the voltage gain, provide more current? should we call it current gain then? or power gain? ^_^
 again different designs might lead to some changes in current, impedance output, etc. so you'd need to know the specifics of a given design or simply measure the output of the amp into a given load. but it's plain wrong to just expect the voltage gain switch to be some magical "turbo" button that boosts everything into betterness. that's just ignorance talking most of the time.
 

last but not least on the web:
 you have probably heard about the psycho acoustic effect of loudness, the famous "louder is better". this has nothing to do with the gear and all to do with the listener. when listening to music, louder(up to a point) feels superior, better bass, better soundstage, clearer details etc. that's how it feels. the average Joe uses his amp, presses the gain switch back and forth and subjectively experiences all those merits of louder sound with higher gain. I'd say most people simply fall for that effect and make up a fallacy involving power as to why it feels better.
to be able to tell anything about the sound without falling victim to the louder is better, we'd need a blind test and matched loudness. something that pretty much no audiophile giving his opinion has ever done. so from an objective point of view, you shouldn't care too much about those feedbacks, and when it comes to power, better trust good old objective electricity laws.

 

[CraftyClown]

Fantastic answer, thank you so much.
 
The world of the Audiophile seems jam packed with myths, hearsay and misinformation, so it's great to have places like this sound science forum to understand how things really work.

 

[castleofargh]

thanks, but keep some of the healthy skepticism that made you come ask a question. even if it's amazing me, who will someday become general of the world, I've been only one voice giving one opinion and not proving much of anything.
maybe nobody else has posted yet because they feel what I posted was ok, but most likely they didn't read because it's too long ^_^.

 

[frigginloony]

  thanks, but keep some of the healthy skepticism that made you come ask a question. even if it's amazing me, who will someday become general of the world, I've been only one voice giving one opinion and not proving much of anything.
maybe nobody else has posted yet because they feel what I posted was ok, but most likely they didn't read because it's too long ^_^.

indeed I have often wondered the same thing. Having multiple cans and multiple amps. I have always noticed that there is the " just perfect volume" precipice  too low   sounds not "full - rich"  too high well I feel that when you jack the sound to just below uncomfortable levels it feels rich but then many cans turn muddled in quality.    however it occurs all over the spectrum depending on the HP and amp combo and even the music doesn't flow with the amp.  It may indeed be my imagination, but i feel that all amps have their "sweet spot" unless they are grossly under powered.  Headroom on a powerful amp gives you the option to model your sound to the headphone, whereas a low or under powered amp may get you there but its just NOT the same. The slam, or punch or tonal richness just isn't quite there.
 
ok now my brain is broke... and it's all Ya'll faults        enough thinking for one day!!   

 

[castleofargh]

ahahah. well about gain switch in some cases it will do no more than help extend the range of the volume knob. you move the knob, reach a limit high or low where it's not possible or not easy to set the volume right, you change the gain and get a all new range to play with and find your sweet spot ^_^. it's at least one of the ways to look at it.
 
 for loudness with gears, I also get the feeling that some devices feel better at a different loudness, but there could be so many reasons, objective or subjective, that I don't know if we can put the blame on something as specific as amp's headroom or gain switch. just finding at which volume level your own ear typically triggers the tightening of the eardrum, and how different signatures trigger it at different levels, could be a vast subject to determine when your ears get the best possible sound and optimal dynamic. then with a different signature/headphone, there is the subjective perceived loudness that could be different from actual loudness. do you use a headphone with a recessed medium at the same loudness as you would a headphone with boosted mids? then for some cans, the distortion numbers might increase to a very audible point at high db, almost factually changing the sound to "mud". maybe the amp struggles because of the headphone's impedance and increasing the loudness just makes that last push that forces the amp into some non linearity(but then the amp was probably not appropriate for that headphone in the first place). maybe it's as I mentioned before, a little channel imbalance in the volume control of your amp and what you consider the sweet spot happens to be a place where the balance is significantly better? so many possible causes that it's unreasonable to draw any kind of conclusion from casual listening of different gears at different levels. but certainly very interesting subjects, all of them if you enjoy torturing your brain some more.

 

 

[CraftyClown]

  thanks, but keep some of the healthy skepticism that made you come ask a question. even if it's amazing me, who will someday become general of the world, I've been only one voice giving one opinion and not proving much of anything.
maybe nobody else has posted yet because they feel what I posted was ok, but most likely they didn't read because it's too long ^_^.

 
You are of course quite right. I was quick to accept your answer as it confirmed my own thoughts
 
As I understand it, this thread from 2010 also addresses the same subject
 
http://www.head-fi.org/t/501017/amplifier-voltage-swing-high-impedance-headphones-sq-difference-proven
 
Unless someone can provide evidence or a compelling argument otherwise, then I'm happy to accept this as yet another audiophile myth

 

[GRUMPYOLDGUY]

You are of course quite right. I was quick to accept your answer as it confirmed my own thoughts

As I understand it, this thread from 2010 also addresses the same subject

http://www.head-fi.org/t/501017/amplifier-voltage-swing-high-impedance-headphones-sq-difference-proven

Unless someone can provide evidence or a compelling argument otherwise, then I'm happy to accept this as yet another audiophile myth

Are you talking about digital gain/attenuation vs analog gain?

Assuming mastering engineers use the full dynamic range allotted to them for the particular format, you want unity gain pre-DAC, then use analog gain to drive the headphones.

If there are unused bits, for example someone stuffed 16 bit data into a 24 bit format with zero padding (on the left!!), you can use some gain pre-DAC to get to full scale without losing any information then use analog gain as necessary.

If the data already uses the full dynamic range and you apply gain pre-DAC, you will lose information.

That's the digital side of things.

You also have to consider hardware performance at various levels. And this is where, as castleofargh says, you would need to make some measurements.

 

[CraftyClown]

Are you talking about digital gain/attenuation vs analog gain?

Assuming mastering engineers use the full dynamic range allotted to them for the particular format, you want unity gain pre-DAC, then use analog gain to drive the headphones.

If there are unused bits, for example someone stuffed 16 bit data into a 24 bit format with zero padding (on the left!!), you can use some gain pre-DAC to get to full scale without losing any information then use analog gain as necessary.

If the data already uses the full dynamic range and you apply gain pre-DAC, you will lose information.

That's the digital side of things.

You also have to consider hardware performance at various levels. And this is where, as castleofargh says, you would need to make some measurements.

 
My original question actually related to the gain settings on my portable DAP the Questyle QP1R that has 3 digital gain settings.
 
I had seen numerous references to people using higher gain, not for volume purposes but because the attached headphones responded better to more power. I didn't understand the logic behind this, hence my questions.

 

[GRUMPYOLDGUY]

My original question actually related to the gain settings on my portable DAP the Questyle QP1R that has 3 digital gain settings.

I had seen numerous references to people using higher gain, not for volume purposes but because the attached headphones responded better to more power. I didn't understand the logic behind this, hence my questions.

Same concept as volume.

It sounds like those people want a different amp. Whether or not they need a different amp is another question. I'm skeptical of their claim. But tk the point of your original question, they could be losing information with those fixed gain settings, especially if volume control on the DAP is already maxed. Presumably they use saturation logic in the DAP.

 

[CraftyClown]

Same concept as volume.

It sounds like those people want a different amp. Whether or not they need a different amp is another question. I'm skeptical of their claim. But tk the point of your original question, they could be losing information with those fixed gain settings, especially if volume control on the DAP is already maxed. Presumably they use saturation logic in the DAP.

 
These examples were with Headphones and IEMs that still had plenty of headroom at the lowest gain setting. They were switching to a higher gain purely because they believed they sounded 'better' this way. My belief was that this would provide no improvement to SQ at equivalent volumes and would in fact be potentially detrimental due to the increased noise floor created by the additional gain.

 

[GRUMPYOLDGUY]

These examples were with Headphones and IEMs that still had plenty of headroom at the lowest gain setting. They were switching to a higher gain purely because they believed they sounded 'better' this way. My belief was that this would provide no improvement to SQ at equivalent volumes and would in fact be potentially detrimental due to the increased noise floor created by the additional gain.

Digital gain doesn't increase the noise floor.

If by headroom, you mean unused MSBs in the DAC word, then it should be fine... Most people don't know if they're saturating/overflowing though, so I'm not sure how they're so certain about it. I'd personally be weary of any gain adjustments and aim for bit perfect.

 

[CraftyClown]

Digital gain doesn't increase the noise floor.

If by headroom, you mean unused MSBs in the DAC word, then it should be fine... Most people don't know if they're saturating/overflowing though, so I'm not sure how they're so certain about it. I'd personally be weary of any gain adjustments and aim for bit perfect.

 
Ok, so this was part of my original question. Coming from a video background where electronic gain is used to boost the sensitivity of the sensor, I am aware that that increase in gain will result in a progressive amount of noise in the resulting image.
 
I presumed this was the same principal with audio devices. If I have misunderstood, then does that mean there is no downside to using more gain and less volume as opposed less gain and more volume on my DAP?

 

[GRUMPYOLDGUY]

Ok, so this was part of my original question. Coming from a video background where electronic gain is used to boost the sensitivity of the sensor, I am aware that that increase in gain will result in a progressive amount of noise in the resulting image.

I presumed this was the same principal with audio devices. If I have misunderstood, then does that mean there is no downside to using more gain and less volume as opposed less gain and more volume on my DAP?

I don't know much about photosensors... Are you talking about this ISO setting (I think it's called)? I have seen that phenomenon with my camera, but I don't know why that happens.

As for audio, see my previous post. More gain/less volume is fine (digitally) for some cases, but will result in a loss of information in other cases... From the analog side you need measurements or detailed specs to make the determination.

 

[CraftyClown]

I don't know much about photosensors... Are you talking about this ISO setting (I think it's called)? I have seen that phenomenon with my camera, but I don't know why that happens.

As for audio, see my previous post. More gain/less volume is fine (digitally) for some cases, but will result in a loss of information in other cases... From the analog side you need measurements or detailed specs to make the determination.

 
That's correct, it is sometimes referred to as ISO. iSO or ASA was the manner in which film sensitivity was measured, the higher the ISO the more sensitive to light the film stock was. The term ISO is still used in Digital cameras, mostly I think to make it easier for those who were used to the old system of sensitivity measurements. The bottom line is that with digital cameras, increasing the ISO/Gain will increase the sensors sensitivity to light.
 
You mention that more gain/less volume is mostly fine, but may result in a loss of information in some cases. Could you briefly explain what the reason for those cases might be?