A vision test for an ideal audiophile:

[Thompsonuxb]

Thomson, you're not only missing the point, you're missing it by a universe sized margin.  The point of level matching is to remove a variable that may, MAY, skew the reviewers perception.  Throwing arguments like 'if one amp is clearly superior to another at any volume' is all well and good, for that amp and the one you're comparing it against although to be really sure you'd have to adjust the levels of respective amps up and down.  Or, you could level match, remove the volume variable and judge their merits.

It has been demonstrated by peer reviewed studies that many people perceive greater volume to be preferential, or to 'sound better' than the same song played at a quieter level. 

No Jota I'm not missing the point - you're refusing to accept the point 'volume' or 'loudness' is not the decider when comparing amps.

(and yes I noticed the MAY)

People all over the world make purchasing judgements based on their own sense - millions of them.

They listen not at fixed levels. I mean Who walks around with 'level' meters in their pockets?

But listen at presentation of the sound, imagery etc - at various levels.

The perceived 'loud sounds better' is rubbish when the fact is a well sorted amp playing at low volumes sounds better than a poor one playing loud .

Point is it makes no sense to compromise it when comparing.

Maybe it's time to review the methods used in those 'peer review studies'

 

[manicm]

Level matching makes no sense to me whatsoever. Some have used a car analogy, but only in context of straight line speed - are you saying a Corolla will tackle a particular bend at 100km/h equally well as a Ferrari? Ha hahahaha.

As far as people preferring to listen at higher volumes - well listen to a reasonably powerful true Class A amp at lower volumes and you'll be shocked.

 

[pauln]

Level matching makes no sense to me whatsoever. Some have used a car analogy, but only in context of straight line speed - are you saying a Corolla will tackle a particular bend at 100km/h equally well as a Ferrari? Ha hahahaha.

As far as people preferring to listen at higher volumes - well listen to a reasonably powerful true Class A amp at lower volumes and you'll be shocked.

So are you saying that Fletcher, Munson, Robinson and Dadson all got it wrong and the equal loudness contours they came up with are wrong? Are you saying that the human ear doesn't in fact work like that? Should we re-write the text books now?

 

[Laurens_B]

Thompson, you really are missing the point, as Jota pointed out. Pauln already warned for this. I'm just glad that he is not qualified to decide on any of these research methodologies.

 

[Thompsonuxb]

Thompson, you really are missing the point, as Jota pointed out. Pauln already warned for this. I'm just glad that he is not qualified to decide on any of these research methodologies.

 

No, I'm not missing the point but I accept we are looking at this from different perspectives.

Level matching at a fixed level is pointless - for starters it's not a realistic view at how amps are used.

Even in a 'controlled' test comparing kit, it makes no sense.

If we are to compare performance, variable levels are required to properly judge how kit performs.

I.e start with the amps output set to 45db@1m with 5db increments upto 95db@1m then you can measure performance - how it sounds at each step.

If one amp tops out at say 65db@1m before clipping..... ref performance you'll have your answer which is better.

Having a 'fixed level' would achieve nothing

In terms of a real world application we will all have our preferred 'level' at which we enjoy our music - Hence just test the amp in question without limiting it to the performance of another.

Point is if we only swap one component be it source, interconnects, speaker cable or speaker they need to be tested/compared at various levels if we are to realistically compare them.

That's all I'm saying.

 

[Thompsonuxb]

To add.

The other week we were talking about compressed and uncompressed music, one member had two copys of a piece of music one compressed the other not, he wanted to compare.

Stated he wanted to level match to compare - I asked why - I did not argue it.

Thing is at lower listening levels the compressed track may well sound better simply because of the compression.

At higher levels the uncompressed version may well sound better because of the better dynamics.

It was never perused - but again that 'detail' would have been lost if the tracks were level matched.

He had both versions a copy to suit him if he was in the mood to play it loud or a mood to play it quite .

 

[Laurens_B]

To add.

The other week we were talking about compressed and uncompressed music, one member had two copys of a piece of music one compressed the other not, he wanted to compare.

Stated he wanted to level match to compare - I asked why - I did not argue it.

Thing is at lower listening levels the compressed track may well sound better simply because of the compression.

At higher levels the uncompressed version may well sound better because of the better dynamics.

It was never perused - but again that 'detail' would have been lost if the tracks were level matched.

He had both versions a copy to suit him if he was in the mood to play it loud or a mood to play it quite .

Nobody has been talking about testing at a fixed level. Of course you can turn them up and one will distort before the other. But when you turn them up, make sure you turn them both up by the same amount, to keep it fair. This is level matching. When one starts to distort (badly), you are not testing within the design parameters anymore (given it's designed competently of course).

Regarding the lossy/lossless test: nothing would have been lost by level matching them. It would have been perfect to compare them that way.

Also, I have not been trying to convince you or anyone about anything regarding amps sound the same or whatever, just trying to explain the logic and reason behind level matching, and why it is perfectly valid in all situations on the issue of finding difference in quality between amps. As with every controlled test environment, change one variable at a time, changing more variables at any instant will make it impossible to know which of the variables caused the change (if any). That's all.

 

[Infiniteloop]

Thompson, you really are missing the point, as Jota pointed out. Pauln already warned for this. I'm just glad that he is not qualified to decide on any of these research methodologies.

No, I'm not missing the point but I accept we are looking at this from different perspectives.

Level matching at a fixed level is pointless - for starters it's not a realistic view at how amps are used.

Even in a 'controlled' test comparing kit, it makes no sense.

If we are to compare performance, variable levels are required to properly judge how kit performs.

I.e start with the amps output set to 45db@1m with 5db increments upto 95db@1m then you can measure performance - how it sounds at each step.

If one amp tops out at say 65db@1m before clipping..... ref performance you'll have your answer which is better.

Having a 'fixed level' would achieve nothing

In terms of a real world application we will all have our preferred 'level' at which we enjoy our music - Hence just test the amp in question without limiting it to the performance of another.

Point is if we only swap one component be it source, interconnects, speaker cable or speaker they need to be tested/compared at various levels if we are to realistically compare them.

That's all I'm saying.

I kinda agree with you about the line level matching thing. For me, it's not about 'loudness' per se, - it's about the sound 'signature' of the Amp being listened to. I know what levels I like to listen to and recently have started monitoring this using an App on my iPhone (not terribly scientific by some standards, but at least it's some sort of reference). With some Amps, you turn them up to a level, and they sound worse (without going anywhere near clipping) and when you turn others up to the same level, they sound simply glorious.

It's about being able to hear into the music. One of my favourite pieces for this kind of comparative listening is Mozart's Gran Partita (Marriner, St Martin in the Fields). The Gran Partita features lots of woodwind instruments which at times sound reedy, resonant and display different timbres depending on the movement. Listening to this piece - (being a favourite, I have heard it many times both live and recorded), and knowing what these instruments are meant to sound like, gives me a good understanding of how the Amp I'm listening to is delivering the music. I'm not interested in specs or distortion figures, I'm only interested in the sound coming from the Amp. - If it delivers what I know to be a realistic reproduction, - I'm happy.

In any event, it's my ears that tell me when something sounds right, not some spec sheet.

 

[pauln]

If we are to compare performance, variable levels are required to properly judge how kit performs.

I.e start with the amps output set to 45db@1m with 5db increments upto 95db@1m then you can measure performance - how it sounds at each step.

If one amp tops out at say 65db@1m before clipping..... ref performance you'll have your answer which is better.

That's all I'm saying.

I can see your point but what you're comparing here is how good the power supply is - how loud can it go before it clips or distorts horribly. You can then say that the amp that plays louder is better in that respect but in order to compare other aspects of an amplifiers performance - i.e. comparing all these audiophile adjectives like soundstage, timbre, inky blackness etc. etc. - you have to have them at the same volume in order to make a valid comparison. Any comparison of things has to be done on a level playing field to be valid. If you were comparing white wines or lagers for instance, you would want them to be at the same temperature - if one was warmer it probably wouldn't taste so good and because they were at different temperatures you wouldn't know if the wine was worse or if it was the warmth making it seem worse. Do you see what I mean? Another example, if you wanted to see which of two cars was fastest you would have to make sure that the weight in each was the same because obviously if one of them had four people onboard and the other had only the driver... you see where I'm going with this?

Then to complicate things further and specifically in terms of audio we have to take into account the well documented and accepted fact that human ears perceive low and high frequencies to be quieter than they are in actual decibels hence the equal loudness contours people talk of. This is why those frequencies are boosted at low volume to make the music seem normal and not bass light. These equal loudness contours flatten as the level increases, so admittedly, at VERY high levels it's less apparent. They will also vary from person to person depending upon how damaged their hearing is but they are always there. This phenomenon has been tested on many, many people and is not a theory or an idea, it's real. Unsurprisingly, humans have the greatest audio acuity in the frequency range of the human voice.

 

[CnoEvil]

FWIW. If I like 2 amps, then it is important that they are level matched, as the loudest one will usually sound better. If I don't like an amp/speaker combination, turning it up just makes it sound worse.

 

[Vladimir]

To add.

The other week we were talking about compressed and uncompressed music, one member had two copys of a piece of music one compressed the other not, he wanted to compare.

Stated he wanted to level match to compare - I asked why - I did not argue it.

Thing is at lower listening levels the compressed track may well sound better simply because of the compression.

At higher levels the uncompressed version may well sound better because of the better dynamics.

It was never perused - but again that 'detail' would have been lost if the tracks were level matched.

He had both versions a copy to suit him if he was in the mood to play it loud or a mood to play it quite .

Logical person:

1) Level match both at the same high/low volume > Listen & Compare > Make valid conclusion about sound quality

2) Set one at high level and other at low level > Listen & Compare > Make valid conclusion about sound quantity[/u]

Thompson:

3) Set one at high level and other at low level > Listen & Compare > Make invalid conclusion about sound quality[/u]

*bye*

 

[Thompsonuxb]

To add.

The other week we were talking about compressed and uncompressed music, one member had two copys of a piece of music one compressed the other not, he wanted to compare.

Stated he wanted to level match to compare - I asked why - I did not argue it.

Thing is at lower listening levels the compressed track may well sound better simply because of the compression.

At higher levels the uncompressed version may well sound better because of the better dynamics.

It was never perused - but again that 'detail' would have been lost if the tracks were level matched.

He had both versions a copy to suit him if he was in the mood to play it loud or a mood to play it quite .

Logical person:

1) Level match both at the same high/low volume > Listen & Compare > Make valid conclusion about sound quality

2) Set one at high level and other at low level > Listen & Compare > Make valid conclusion about sound quantity

Thompson:

3) Set one at high level and other at low level > Listen & Compare > Make invalid conclusion about sound quality

*bye*

Lol..... Keep up man.

 

[manicm]

I still maintain level matching is patent nonsense. A good, say 100w, amplifier will have a more robust and bigger power supply than a 30w amp, and even at the same db level it's going to consume more power unless a very clever design is implemented. So you're comparing apples with oranges really. And if you're using the same speakers, which you should if you're doing this exercise, I doubt the results will be the same.

 

[Vladimir]

To add.

The other week we were talking about compressed and uncompressed music, one member had two copys of a piece of music one compressed the other not, he wanted to compare.

Stated he wanted to level match to compare - I asked why - I did not argue it.

Thing is at lower listening levels the compressed track may well sound better simply because of the compression.

At higher levels the uncompressed version may well sound better because of the better dynamics.

It was never perused - but again that 'detail' would have been lost if the tracks were level matched.

He had both versions a copy to suit him if he was in the mood to play it loud or a mood to play it quite .

Logical person:

1) Level match both at the same high/low volume > Listen & Compare > Make valid conclusion about sound quality

2) Set one at high level and other at low level > Listen & Compare > Make valid conclusion about sound quantity

Thompson:

3) Set one at high level and other at low level > Listen & Compare > Make invalid conclusion about sound quality

*bye*

Lol..... Keep up man.

I listed, underlined and bolded! That makes it facts!

Anyways. Checkout this TV show sitcom when you are bored. Fresh off the boat.

 

[Thompsonuxb]

To add.

The other week we were talking about compressed and uncompressed music, one member had two copys of a piece of music one compressed the other not, he wanted to compare.

Stated he wanted to level match to compare - I asked why - I did not argue it.

Thing is at lower listening levels the compressed track may well sound better simply because of the compression.

At higher levels the uncompressed version may well sound better because of the better dynamics.

It was never perused - but again that 'detail' would have been lost if the tracks were level matched.

He had both versions a copy to suit him if he was in the mood to play it loud or a mood to play it quite .

Logical person:

1) Level match both at the same high/low volume > Listen & Compare > Make valid conclusion about sound quality

2) Set one at high level and other at low level > Listen & Compare > Make valid conclusion about sound quantity

Thompson:

3) Set one at high level and other at low level > Listen & Compare > Make invalid conclusion about sound quality

*bye*

Lol..... Keep up man.

I listed, underlined and bolded! That makes it facts!

Anyways. Checkout this TV show sitcom when you are bored. Fresh off the boat.

You either you don't understand what's being said or you do and now intend to derail the thread with pointless images and links....

 

[Vladimir]

The show is funny-ish and has urban tunes.

 

[Thompsonuxb]

Thomson, you're missing the point...

Listen, and understand. That's what Thompson does. That's all he does. He can't be bargained with. He can't be reasoned with. He doesn't feel pity, or remorse, or fear. And he absolutely will not understand, ever, until you are dead.

Steve1979 too, maybe the penny is dropping..... Lol.

 

[Vladimir]

Thomson, you're missing the point...

Listen, and understand. That's what Thompson does. That's all he does. He can't be bargained with. He can't be reasoned with. He doesn't feel pity, or remorse, or fear. And he absolutely will not understand, ever, until you are dead.

Ahhaahhaahhaha! *ROFL*

 

[steve_1979]

Steve1979 too, maybe the penny is dropping..... Lol.

Sorry Thompson. True a word said in jest and all that but I'm only pulling you leg really.

You're a friendly chap and this forum would be a far duller place without you. *drinks*

 

[Vladimir]

Affirmative.

 

[Thompsonuxb]

To add.

The other week we were talking about compressed and uncompressed music, one member had two copys of a piece of music one compressed the other not, he wanted to compare.

Stated he wanted to level match to compare - I asked why - I did not argue it.

Thing is at lower listening levels the compressed track may well sound better simply because of the compression.

At higher levels the uncompressed version may well sound better because of the better dynamics.

It was never perused - but again that 'detail' would have been lost if the tracks were level matched.

He had both versions a copy to suit him if he was in the mood to play it loud or a mood to play it quite .

Nobody has been talking about testing at a fixed level. Of course you can turn them up and one will distort before the other. But when you turn them up, make sure you turn them both up by the same amount, to keep it fair. This is level matching. When one starts to distort (badly), you are not testing within the design parameters anymore (given it's designed competently of course).

Regarding the lossy/lossless test: nothing would have been lost by level matching them. It would have been perfect to compare them that way.?

Also, I have not been trying to convince you or anyone about anything regarding amps sound the same or whatever, just trying to explain the logic and reason behind level matching, and why it is perfectly valid in all situations on the issue of finding difference in quality between amps. As with every controlled test environment, change one variable at a time, changing more variables at any instant will make it impossible to know which of the variables caused the change (if any). That's all.

 

Sounds like we are getting there.....

But has I've said no point in compromising the amp that can keep going..... :-D

 

[Thompsonuxb]

If we are to compare performance, variable levels are required to properly judge how kit performs.

I.e start with the amps output set to 45db@1m with 5db increments upto 95db@1m then you can measure performance - how it sounds at each step.

If one amp tops out at say 65db@1m before clipping..... ref performance you'll have your answer which is better.

That's all I'm saying.

I can see your point but what you're comparing here is how good the power supply is - how loud can it go before it clips or distorts horribly. You can then say that the amp that plays louder is better in that respect but in order to compare other aspects of an amplifiers performance - i.e. comparing all these audiophile adjectives like soundstage, timbre, inky blackness etc. etc. - you have to have them at the same volume in order to make a valid comparison. Any comparison of things has to be done on a level playing field to be valid. If you were comparing white wines or lagers for instance, you would want them to be at the same temperature - if one was warmer it probably wouldn't taste so good and because they were at different temperatures you wouldn't know if the wine was worse or if it was the warmth making it seem worse. Do you see what I mean? Another example, if you wanted to see which of two cars was fastest you would have to make sure that the weight in each was the same because obviously if one of them had four people onboard and the other had only the driver... you see where I'm going with this?

Then to complicate things further and specifically in terms of audio we have to take into account the well documented and accepted fact that human ears perceive low and high frequencies to be quieter than they are in actual decibels hence the equal loudness contours people talk of. This is why those frequencies are boosted at low volume to make the music seem normal and not bass light. These equal loudness contours flatten as the level increases, so admittedly, at VERY high levels it's less apparent. They will also vary from person to person depending upon how damaged their hearing is but they are always there. This phenomenon has been tested on many, many people and is not a theory or an idea, it's real. Unsurprisingly, humans have the greatest audio acuity in the frequency range of the human voice.

I've always maintained an amp is built to spec. We cannot ignore its power supply or any other aspect of its design.

Are we almost agreeing?

 

[steve_1979]

Thomson, you're missing the point...

That's what Thomson does. That's all he does. He can't be bargained with. He can't be reasoned with. He doesn't feel pity, or remorse, or fear. And he absolutely will not understand, ever, until you are dead.

 

[pauln]

If we are to compare performance, variable levels are required to properly judge how kit performs.

I.e start with the amps output set to 45db@1m with 5db increments upto 95db@1m then you can measure performance - how it sounds at each step.

If one amp tops out at say 65db@1m before clipping..... ref performance you'll have your answer which is better.

That's all I'm saying.

I can see your point but what you're comparing here is how good the power supply is - how loud can it go before it clips or distorts horribly. You can then say that the amp that plays louder is better in that respect but in order to compare other aspects of an amplifiers performance - i.e. comparing all these audiophile adjectives like soundstage, timbre, inky blackness etc. etc. - you have to have them at the same volume in order to make a valid comparison. Any comparison of things has to be done on a level playing field to be valid. If you were comparing white wines or lagers for instance, you would want them to be at the same temperature - if one was warmer it probably wouldn't taste so good and because they were at different temperatures you wouldn't know if the wine was worse or if it was the warmth making it seem worse. Do you see what I mean? Another example, if you wanted to see which of two cars was fastest you would have to make sure that the weight in each was the same because obviously if one of them had four people onboard and the other had only the driver... you see where I'm going with this?

Then to complicate things further and specifically in terms of audio we have to take into account the well documented and accepted fact that human ears perceive low and high frequencies to be quieter than they are in actual decibels hence the equal loudness contours people talk of. This is why those frequencies are boosted at low volume to make the music seem normal and not bass light. These equal loudness contours flatten as the level increases, so admittedly, at VERY high levels it's less apparent. They will also vary from person to person depending upon how damaged their hearing is but they are always there. This phenomenon has been tested on many, many people and is not a theory or an idea, it's real. Unsurprisingly, humans have the greatest audio acuity in the frequency range of the human voice.

I've always maintained an amp is built to spec. We cannot ignore its power supply or any other aspect of its design.

Are we almost agreeing?

Light years away.