The Audio Path In Consumer-Grade Products

[<eargiant]

Gain staging is not some unknown mystery. For the sake of maximizing the s/n and reducing distortion, you want your line level and/or pre amp stages to run hot as possible while maintaining enough headroom to avoiding clipping, and subsequently using only as much gain from the amp as needed to achieve desired listening levels. Proper gain structure will ensure plenty of headroom and should keep any residual noise buried well below audible levels.

Remember, in my post that you quoted I was responding to a member that asked "How is reducing the gain different from reducing the volume?"

No one is implying that it is a dark art, only that it is one simple tweak in the audio path that designers can make that will clearly alter the sound a unit produces. In some cases designers opt for a higher gain because it will give the impression that an amp is more powerful when the volume knob is turned. It will also sound more forward and impactful.

So it's not that it's a mystery but more so that many enthusiasts are not even aware of it - otherwise I wouldn't hear the line that everything sounds the same so often.

What home audio pre-amp do you have that has a gain knob?

That's a good point, many don't offer an option so the user only gets the choice the designer made and it's not always for the best reason.

I don't know about his amp but my integrated has a "Jump" switch that is specifically there to allow you to bypass the gain of the flat amp. Mind you, this amp has no tone controls so it is already "direct".

The difference in sound presentation without the additional gain is undeniable. It's like having two different amps in one.

 

[PENG]

otherwise I wouldn't hear the line that everything sounds the same so often.

I don't understand why you are still so fixated on this point that seems to be moot. I have yet to see evidence that people on this site have said such a thing. I mean, "everything sounds the same", who the heck would say such a thing? Certainly not Goliath, and I thought we are audioholics afterall, even Goliath. I have seen enough of saying such as "all well designed amps should sound the same if used within their limits". I don't fully agree to such saying either, but for someone trying to make a point under context, I can almost accept that, though that is still a far cry from what you seemed to be quoting. I guess I am repeating things I said before, but I thought it is important that visitors to this site don't get the wrong impression.

 

[M Code]

Component gain is a critical factor in pro-audio applications, since there is a very wide diversity in component specifications for input/output for levels and impedances. Regarding consumer products there are fewer variables, they just want plug/play and go.

Key point for a proper gain setting is to have its input sensitive enough to swing the component into it sweet output range..
If too low the noise floor jumps up...
If too high it will be overdriven into clipping.
With the older analog components it was less critical as their dynamic range was much lower, but now with digital, uncompressed streams the possibility of overdriving and clipping is far more crucial.

Back in the early 80s we worked for a major audio brand that co-invented the CD, and the 1st CDs were of fairly low levels for the mixing engineers feared overdriving the signal and when a digital stream clips it is far more destructive than analog clipping. But in the following years of later 80s, there were multiple major developments for amplifiers and loudspeaker capable of handling the higher dynamic range.

Just my $0.02...

 

[lovinthehd]

Remember, in my post that you quoted I was responding to a member that asked "How is reducing the gain different from reducing the volume?"

No one is implying that it is a dark art, only that it is one simple tweak in the audio path that designers can make that will clearly alter the sound a unit produces. In some cases designers opt for a higher gain because it will give the impression that an amp is more powerful when the volume knob is turned. It will also sound more forward and impactful.

So it's not that it's a mystery but more so that many enthusiasts are not even aware of it - otherwise I wouldn't hear the line that everything sounds the same so often.



That's a good point, many don't offer an option so the user only gets the choice the designer made and it's not always for the best reason.

I don't know about his amp but my integrated has a "Jump" switch that is specifically there to allow you to bypass the gain of the flat amp. Mind you, this amp has no tone controls so it is already "direct".

The difference in sound presentation without the additional gain is undeniable. It's like having two different amps in one.

What integrated amp is it?

 

[killdozzer]

What home audio pre-amp do you

have that has a gain knob?

Hi, lovinthehd!

There's still room for me to be corrected, but let me try and set the record straight.

This wasn't mine, it was my father's amp, an old NAD THX Lucasfilm:

It was a preamp and amp separates. You could turn up the input level. My father used this amp to run a pair of these:

I circled the thing I aim at, the knobs aren't labelled "gain" and that's why I say you still get to correct me if necessary.

The effect was this; turning these knobs up would result in a sound that is full of midds, very pronounced and present (almost as if compressed) and if you would so much as touch the master volume you'd get a high SPL. If you turned them all the way down, the overall sound would be deeper, softer, the bass fuller and you would have to turn the master volume knob to three quarters to get the same SPL.

The important thing is it would affect the final sound even at same volume.

 

[RichB]

I feel wholly unfit to post on this thread because I am not a EE and have actually listened to the AHB2 in my system.

FWIW: GSR brought over his AHB2 over lunch so we had limited time. Single channel, the high-end sounded excellent but it did not seem to drive the bass that well. I expected bi-amping to be an improvement but didn't hear one. When bridged, it sounded more balanced with no change to the high-end. Perhaps the gain changes, I don't know but I would bridge this amp with the Salon2's.

- Rich

 

[killdozzer]

@lovinthehd

Further research lead me to these explanations in the owners manual.

 

[PENG]

Component gain is a critical factor in pro-audio applications, since there is a very wide diversity in component specifications for input/output for levels and impedances. Regarding consumer products there are fewer variables, they just want plug/play and go.

Key point for a proper gain setting is to have its input sensitive enough to swing the component into it sweet output range..
If too low the noise floor jumps up...
If too high it will be overdriven into clipping.
With the older analog components it was less critical as their dynamic range was much lower, but now with digital, uncompressed streams the possibility of overdriving and clipping is far more crucial.

Back in the early 80s we worked for a major audio brand that co-invented the CD, and the 1st CDs were of fairly low levels for the mixing engineers feared overdriving the signal and when a digital stream clips it is far more destructive than analog clipping. But in the following years of later 80s, there were multiple major developments for amplifiers and loudspeaker capable of handling the higher dynamic range.

Just my $0.02...

Improper gain structure is the real reason why changing the gain of one component may change the sound due to distortions resulted from one of more components get over driven, as you said, to outside of a component's sweet range or even clipping resulted in audible distortions. If all components remain within their "sweet" operating range, changing the gain should not change the sound as there will be no audible change in the distortion level.

 

[<eargiant]

I don't understand why you are still so fixated on this point that seems to be moot. I have yet to see evidence that people on this site have said such a thing. I mean, "everything sounds the same", who the heck would say such a thing? Certainly not Goliath, and I thought we are audioholics afterall, even Goliath. I have seen enough of saying such as "all well designed amps should sound the same if used within their limits". I don't fully agree to such saying either, but for someone trying to make a point under context, I can almost accept that, though that is still a far cry from what you seemed to be quoting. I guess I am repeating things I said before, but I thought it is important that visitors to this site don't get the wrong impression.

Remember, the last time you incredulously asked me for proof about the DAC comment I referenced - I provided it for you . I think you just like making me work .

Anyway, it wasn't a "quote" as you say, I was paraphrasing and I stand by it. When I started reading posts here on AH I'd come across so many of the - just buy this AVR vs (insert brand here) integrated because you won't get better sound comments that I was in disbelief. How did they know the engineers made the same design choices? I'll accept that if they are talking about models within the D&M line-up but otherwise- no dice. It was like people were selecting the science they wanted to believe and dismissing the science they didn't care for. Gain, bias, capacitors (size, quantity & location), design, signal path, heatsinks, quality & quantity of outputs, filter capacitance, transformers, etc., etc., is considered "over-engineering" by many here as long as we can get good measurements without that stuff.

I think if I gave some here just one cheap output transistor with a dab of grease strapped to tin foil in a box it would be fine as long as I could make it measure OK with a test tone for a few minutes. That's not progress, that's a race to the bottom. In Gene's video review of the Yamaha A-S801 he and Hugo made a BIG deal about the construction of that unit. I have one and I think that should be a minimum for audio gear. Some of you would call that over-engineered, I call it properly designed.

I almost bailed on this place a while back (I think some would be happy if I did ) but I think there's good stuff here. BTW, I came because I do NOT buy into the cable BS that is out there.

I'm glad you mentioned the line that "all well designed amps should sound the same if used within their limits" and the fact that you don't fully agree with that. I definitely disagree with that quote especially when one is not discussing separates. Usually I hear people make that claim or something to the effect that if it sounds different something is wrong with it. This is where we diverge and maybe why I don't think it's a "moot" point. Two amps CAN be well designed, not distort, measure beautifully and sound different due to small differences like gain, biasing, etc.. I think there are many here that don't believe that otherwise I wouldn't read some of the recommendation comments I do.

Either way, it seems like you and I agree more than we disagree.

Improper gain structure is the real reason why changing the gain of one component may change the sound due to distortions resulted from one of more components get over driven, as you said, to outside of a component's sweet range or even clipping resulted in audible distortions. If all components remain within their "sweet" operating range, changing the gain should not change the sound as there will be no audible change in the distortion level.

Unfortunately, I strongly disagree with this statement. As proof I offer the amp with the "Jump" switch in the picture I posted earlier. I can assure you that the designer did NOT put that switch there to offer an "improper gain structure" option to change the sound. They were striving for the proverbial "wire with gain". I believe this manufacturer only had that switch on their top 2 models. Why would they tarnish their no holds barred statement pieces with an "improper gain structure"?

Weather the switch is on or off, this amp will measure beautifully in all parameters and still have extremely low distortion numbers which fall into the "inaudible" category so I can guarantee you that your opinion on why it may sound different is incorrect in this case.

An "improper" design can not perform like this @ 60 kHz.

 

[PENG]

I have never said anything about whatever amp you referred to, that has anything improper, so please do not put words in me. If you are not clear about what I said, please re-read my post instead of just quoting. The paragraph you quoted me had nothing to do with the amp you are now referring to and was not directed to you at all so you are mixing things up.

I did say that if a change in gain does not result in the component being driven outside of its sweet operating range then it should not sound audibly different. As far as I am concerned I am not giving an opinion but making a generic statement of fact in response to MCode's post#263, not yours. You can disagree all you want and there is nothing I can do about it.

As you said, we agreed on certain things. Fortunately we don't have to agree on everything. At this point it is better we stopped going back on forth on something that appears to be going no where, so let's move on. Now I am going back to enjoy my gear and music, the only things that are important in the end.

 

[<eargiant]

I have never said anything about whatever amp you referred to, that has anything improper, so please do not put words in me. If you are not clear about what I said, please re-read my post instead of just quoting. The paragraph you quoted me had nothing to do with the amp you are now referring to and was not directed to you at all so you are mixing things up.

I did say that if a change in gain does not result in the component being driven outside of its sweet operating range then it should not sound audibly different. As far as I am concerned I am not giving an opinion but making a generic statement of fact in response to MCode's post#263, not yours. You can disagree all you want and there is nothing I can do about it.

As you said, we agreed on certain things. Fortunately we don't have to agree on everything. At this point it is better we stopped going back on forth on something that appears to be going no where, so let's move on. Now I am going back to enjoy my gear and music, the only things that are important in the end.

First, for clarity let's state the obvious. If an amp is distorting it will sound different.

PENG, we may not have to agree but we should agree on facts. I can't let this one go because this is the kind generalization that feeds those that think if a unit is designed and operating within the a proper operating range you won't hear a difference.

That's a broad brush you're painting with when you told M Code that "If all components remain within their "sweet" operating range, changing the gain should not change the sound as there will be no audible change in the distortion level." Especially after you just tied the sound difference that may be audible to distortion.

That is factually incorrect.

This is from the user manual of the amp with the "Jump" switch.

"The tonal quality of reproduction can change much depending on the particular signal path through which the signal flows. With your unit you can select, by using the JUMP switch, from two different signal paths as follows...You can bypass the flat amp: set the JUMP switch to -14dB. You can enjoy the purest sound reproduction with the shortest signal path".

Again, based on your logic, this amp must still be within the "sweet-spot" because there is no distortion. Yet the sound is different. Remember, this is not bypassing tone controls as this amp has none. So in essence the options are "pure-direct" and "pure direct with less gain".

I can tell you that the audible difference is far greater than the difference one hears when selecting the typical "pure direct" switch to bypass the tone control signal path on units that have them.

 

[killdozzer]

That is factually incorrect.

This is from the user manual of the amp with the "Jump" switch.

From your post. Read this again. To state something is factually incorrect and then back that with a citation from a manual of the producer who is trying to sell (and possibly overcharge) a unit, is not something you're suppose to be pleased with.

Some conclusions may be drawn from this, but I'm afraid you'd enjoy very few of them. One might say you're gullible when it comes to marketing content and you couldn't say he is wrong about that.

To even suggest that a manual makes for a sound argument would have to imply that whenever there is a weakness in a unit, the manual clearly states it and, thus, since your manual says this is something great, it has to be great.

All the while, we live in a world where no manual ever directly stated that there's something wrong with the product it refers to. And even when there is, it is extremely carefully disguised as an advice to not use a unit in this way or that way because it wasn't meant for that.

 

[Goliath]

I'm glad you mentioned the line that "all well designed amps should sound the same if used within their limits" and the fact that you don't fully agree with that. I definitely disagree with that quote especially when one is not discussing separates.

Okay, so where is the real world evidence that well designed amplifiers sound different if used within their capabilities (ie not clipping), carefully voltage-matched (same volume levels) under double-blind conditions (for parity's sake)?

Please show us examples (make, model) with corresponding audibility data confirming the above.

Usually I hear people make that claim or something to the effect that if it sounds different something is wrong with it. This is where we diverge and maybe why I don't think it's a "moot" point. Two amps CAN be well designed, not distort, measure beautifully and sound different due to small differences like gain, biasing, etc..

Okay, so that is your "hypothesis".

Now, again, where is the real-world evidence that two well designed amps, not distorting, used in identical conditions (level-matched and under double blind conditions) do sound different? Not can, in some hypothetical universe, but do in the real world.

If there are differences in gain, that will be accounted for as both amplifiers must be playing at the same voltage level. We are assuming here that no switches are engaged that are necessarily affecting either the volume, or somehow modifying the frequency response in any way, because that would violate the definition of an "amplifier".

All we need to support this hypothesis are examples and audibility studies to go along with it. Look forward to seeing this data soon.

I think there are many here that don't believe that otherwise I wouldn't read some of the recommendation comments I do.

Correct, and there is good reason not to believe it ...

So the same deal applies above. If you have reliable evidence to support your idea/theory/belief then now is the time to present it. I'm sure my Audioholic peers are very excited to review this data if and when it is submitted.

 

[Goliath]

First, for clarity let's state the obvious. If an amp is distorting it will sound different.

That's a broad brush you're painting with when you told M Code that "If all components remain within their "sweet" operating range, changing the gain should not change the sound as there will be no audible change in the distortion level." Especially after you just tied the sound difference that may be audible to distortion.

That is factually incorrect.

This is from the user manual of the amp with the "Jump" switch.

Sure, but there is no evidence that the switch "which claims to improve signal purity or whatever" actually does improve signal purity in any audible way. Since audibility is all that actually matters.

 

[Steve81]

Now, again, where is the real-world evidence that two well designed amps, not distorting, used in identical conditions (level-matched and under double blind conditions) do sound different? Not can, in some hypothetical universe, but do in the real world.

One should keep in mind that well-designed doesn't necessarily mean well-implemented or operating perfectly to spec in the real world. I've owned an amp (Outlaw 7125) that was distinctly recognizable at lower levels due to transformer hum and excessive noise from some channels, and those issues continued even after it had been returned for servicing. The Marantz that replaced it and the Onkyo AVR that preceeded it never had any of those issues. I know some other members here have reported similar issues with other amps as well.

 

[RichB]

Here is what we know:

• There have been amplifiers that measured well but driving speakers had issues. For example, the first Pioneer ICE receiver has issues driving 4 ohm loads that was discovered by AH.
• Amplifiers are not well tested into reactive loads. For example, HomeTheaterHifi does this occasionally for massive monoblock maps but never for an AVR.
• It is not known at what point an amplifier is not operating within "it's limits" driving a reactive load.
• There is no "standard" speaker load
• Some amplifier distortion measurements show significant increases in distortion when tested with mutli-tone tests. Some don't.
• Soft-clipping is not considered distortion, but it really is. It is difficult to hear sound that is not produced.

- Rich

 

[Pogre]

If you have reliable evidence to support your idea/theory/belief then now is the time to present it. I'm sure my Audioholic peers are very excited to review this data if and when it is submitted.

"Very excited" isn't an exaggeration either. I would love to review data like this as it would be the first piece of evidence I've ever seen to support the hypothesis. Problem is, I don't think it exists.

 

[Goliath]

First, for clarity let's state the obvious. If an amp is distorting it will sound different.

PENG, we may not have to agree but we should agree on facts. I can't let this one go because this is the kind generalization that feeds those that think if a unit is designed and operating within the a proper operating range you won't hear a difference.

To be clear, this is a straw man position of your own making.

If an amp is operating within its proper operating range (that is, if it is operating safely below it's rated power) said amplifier can still sound different to another amplifier operating within it's proper operating range.

The reasons why are well known and established, so I'll point them out to you again so that there is no further confusion :

1) Mismatched volume levels
2) Sighted bias
3) High output impedance leading to irregular frequency response shaping

Proper operating range does *nothing* to prevent the above 3 qualifiers from messing up the amplifiers performance to such an extent that the sound becomes audible to ears.

Really, you only need to screw up 1) and you'll have audible differences all day long. 2) speaks for itself and 3) an amplifier ideally should have a low output impedance or be otherwise load intolerant, therefore a high output impedance destroys any case of the amplifier behaving like an "amplifier" as opposed to a non-defeatable EQ box whose effects vary depending on the speakers connected to it.

So back to this business of amplifiers sounding different even if they fit all the criteria above. Where is the evidence that two amplifiers connected to a particular load sound different if both have low output impedance, are voltage-matched and are both safely operating below their maximum output powers in double blind conditions?

Because that's your contention. So, where is the meat and potatoes to go along with that claim?

 

[AcuDefTechGuy]

...where is the meat and potatoes...

You're making me hungry.

Is it dinner time yet?

 

[AcuDefTechGuy]

I know that John and Jeff's take on AVR sound quality (even TOTL models) is highly contested on Audioholics but since most here seem to rely on science and measurable facts, I was wondering if anyone was in disagreement with the facts they present. Is anyone swayed by this or are those guys just full of it?

I forget. What is this point of your thread?

Using separates with higher quality and more expensive parts and better design will sound significantly unequivocally night-and-day better than a pre-pro like the $4,000 Marantz AV-8802A and all the AVRs out there?

Or using an external DAC will produce sound that is significantly unequivocally night-and-day better the DACs inside a pre-pro like the $4,000 Marantz AV-8802A and all the AVRs out there?