Does It Really Sound The Same?

[bigshot]

These are the kinds of things that are often 40-60dB down relative to the primary signal, and are also, as I said earlier, exactly the data that is tossed in lossy compression.

Considering normal people don't listen to music much louder than 40-60dB, I'd say that kind of thing isn't worth worrying about.

 

[bigshot]

I keep saying this, but here goes again...

Home audio electronics are *designed* to all sound the same... flat response, free of audible distortion, accurate dynamics. It is *calibrated* to the exact same specifications. If it truly does sound different, either some component is failing or there is a serious design flaw. I've had fifty dollar cd players that produced perfect redbook spec sound. Every amp I have ever owned for the past 20 years has done the same. It isn't too much to expect a low power headphone amp or DAC to do the same.

Every pair of speakers I have ever heard has sounded different. Some sound absolutely abysmal. I've found huge variation in headphones too. Some sound great and others induce headaches. There is the end of the animal where people should Be doing listening tests. It would do some good there. Splitting fractions down to completely inaudible levels and then arguing over whether we can hear the unhearable is silly.

 

[fishski13]

Quote:

 


 
These anecdotes sadly lack serious rigor in the administration and recording department. For starters the subject ***always*** knows that something has been changed, you can easily more hear a change when you know a change has been made, - when folks do cheats amazingly people hear massive differences between A and A, secondly these tests seldom randomise the order of presentation or require that the subject repeat the analysis without foreknowledge of changes. The subject is never required to identify the different items blind having heard them once. Then there is the Clever Hans effect - testers frequently subconciously (Being generous)  prime subjects to react to stimulae, also someone says something and we reinterpret it to meet our own beliefs - the language of audiophiles makes this almost inevitable. Since opamps are by definition gain devices then simply not ensuring that the levels are matched renders comparisons invalid even assuming you can accurately hold the memory of a musical rendering that may be 5 or 10 minutes prior which is very difficult unless you are comparing Led Zeppelin with Mahler.
 
If opamps are audibly different (which is highly possible) then that difference should survive much more rigorous testing, for instance record some samples with different opamps in circuit, trim, align and level match and then blind test - this is not rocket science - make samples available for others to audition - seriously it is not difficult - it is a bit time-consuming 
 
 
 
 

hey Nick, you still own that audiophoolery piece of M3?  have you tried rolling opamps?  unlike cables, DBT opamps is a bit more work.

 

[mrspeakers]

Quote:

Considering normal people don't listen to music much louder than 40-60dB, I'd say that kind of thing isn't worth worrying about.

Wow, the perfect blend of totally dismissive and utterly uninformed.  
 
Multiple surveys over the last twenty years have shown people listening to personal music players (from Walkman to today's systems) listen in a range that typically varies from 56 to more than 110dB, with average users listening in quiet environments at 85dB.  This shows that a) a good number of people are punching holes in their eardrums, and b) certain people on this thread have big opinions and a small grasp of facts.
 
Certain people on this thread have extremely strong opinions and a very questionable grasp of engineering or acoustics/psychoacoustics.  As such, I'll bid this thread a "adios," which is a pity as it could have been interesting had people had open minds instead of soap boxes.
 
 

 

[nick_charles]

Quote:

hey Nick, you still own that audiophoolery piece of M3?  have you tried rolling opamps?  unlike cables, DBT opamps is a bit more work.

I still have the M^3 , I have never rolled opamps since I had it built specifically with an opamp that I had good experiences with in a different amp (Pimeta)  - I was a true believer back then and inertia has made me keep it.
 
Reading NWAvguy's review of the mini M3 I was a bit concerned about the terrible crosstalk  (relatively speaking)  and if it affected the full sized M^3 - ihn which case I would have sold it and bought another Zero - however I tested how audible this was (to me) by taking a track, deleting the right channel unplugging the left side of my HD580s and playing it back. At full volume I can just about make out above the noise (on a fairly loud track - The Supermen (David Bowie) ) the bleed-through from left to right - with nothing in the left transducer, at 12 O'clock it is very nearly undetectable (again with zero input on the left) - I generally listen at about 9 O'Clock so it is not worth worrying about...
 
 
 
 

 

[jcx]

L,R channel crosstalk in headphone amps is pretty much just an irrelevant "numbers game" - phonograph playback probably never achieves even 40 dB of channel separation - the industry early on determined that below ~ -20 dB crosstalk was inaudible/not objectionable with music
 
the mini^3 and other "3-channel" implementations are slightly worse on crosstalk than simpler schemes with good gnd layout - but at levels that really shouldn't be expected to be audible - the TRS jack/plug common gnd contact R is expected to be a bigger crosstalk causing error term than the amp gnd topology in portables using 1/8" TRS connectors
 
you need get to 1/4" Ag contact Neutrik TRS connectors to be able to  really worry about amplifier/gnd crosstalk - or rewire with 4-pin "balanced" connectors
 
 
my objection to "3-channel" is purely engineering based - it doesn't "fix" the "errors" it is claimed to - not that I expect it to be audible
 
 

 

[fishski13]

Quote:

I still have the M^3 , I have never rolled opamps since I had it built specifically with an opamp that I had good experiences with in a different amp (Pimeta)  - I was a true believer back then and inertia has made me keep it.
 
Reading NWAvguy's review of the mini M3 I was a bit concerned about the terrible crosstalk  (relatively speaking)  and if it affected the full sized M^3 - ihn which case I would have sold it and bought another Zero - however I tested how audible this was (to me) by taking a track, deleting the right channel unplugging the left side of my HD580s and playing it back. At full volume I can just about make out above the noise (on a fairly loud track - The Supermen (David Bowie) ) the bleed-through from left to right - with nothing in the left transducer, at 12 O'clock it is very nearly undetectable (again with zero input on the left) - I generally listen at about 9 O'Clock so it is not worth worrying about...
 
 
 
 

the BM DAC1 partnered with the M3 is a microscope to the recording, emphasizing the recording over the performance.
 
i would take NWAvguy's Mini-3 measurements with a big grain of salt.

 

[bigshot]

Multiple surveys over the last twenty years have shown people listening to personal music players (from Walkman to today's systems) listen in a range that typically varies from 56 to more than 110dB, with average users listening in quiet environments 85dB.

110 dB is cranking a digital player full blast with mp3s slamming into the edge of clipping. That's about as loud as a chainsaw. If I remember correctly, 85dB is the loudest safe volume for extended listening, and that is even pushing it according to some folks.

You may be correct that many people are destroying their hearing by listening at ridiculous volume levels, but a year or two of that will raise the noise floor on their tinnitus to well above the -40 to 60dB harmonic distortion you're talking about.

Someone listening to music at 110dB isn't a golden eared audiophile (for very long...)

 

[upstateguy]

Quote:

I would not be terribly shocked if the numbers on the M^3 turned out make up, calculated on the back of an envelope, fudged, or otherwise wrong...
 
By virtue of the interchannel IMD that the M^3 probably has, I would be surprised if the GS-1 was measurably better when properly tested.

And you believe nwavguy's measurements and blog because??????    Because someone else has reproduced his findings??????
 


Quote:

 
<snip>
 
If opamps are audibly different (which is highly possible) then that difference should survive much more rigorous testing, for instance record some samples with different opamps in circuit, trim, align and level match and then blind test - this is not rocket science - make samples available for others to audition - seriously it is not difficult - it is a bit time-consuming 
 

Hey Nicky, the claim is that all opamps sound the same, that all DACs sound the same and that all amps sound the same.................... I agree that it's not rocket science to run your tests.....  would you care to have at it????? 
 
Here, in the back of the bus, we are allowed to question broad, sweeping, generalized statements and ask for proof.
 
Would any of the guys who are busy making those claims care to offer the same kind of proof we routinely ask the cable believers for???
 
I don't have the answers, so I'd love to see the proof, or even some of the proof, or even an attempt at the proof.  (diffMaker)
 

 

[Willakan]

@mrspeakers
Just to say, I'm not saying that all opamps sound the same. I don't know where people keep getting the "all" bit from. However, when someone says that something that measures exceptionally well is "tainted" by using cheap opamps and that they can distinguish that in the signal by ear, I am extremely sceptical.
 
Also, I don't give a proverbial how long you've been in the industry, if you say you can hear things which are not shown by null tests I call BS. What exactly could not be shown in a null test? It wouldn't matter if it was 40-60db below the output level - that would be clearly seen in a null difference test.
 
 

 

[upstateguy]

 
Quote:

@mrspeakers
Just to say, I'm not saying that all opamps sound the same. I don't know where people keep getting the "all" bit from. However, when someone says that something that measures exceptionally well is "tainted" by using cheap opamps and that they can distinguish that in the signal by ear, I am extremely sceptical.
 
Also, I don't give a proverbial how long you've been in the industry, if you say you can hear things which are not shown by null tests I call BS. What exactly could not be shown in a null test? It wouldn't matter if it was 40-60db below the output level - that would be clearly seen in a null difference test.
 
 

OK Willie, for the sake of clarity then, what is your position?
 
 
 

 

[nick_charles]

Quote:

Hey Nicky, the claim is that all opamps sound the same, that all DACs sound the same and that all amps sound the same.................... I agree that it's not rocket science to run your tests.....  would you care to have at it????? 

Sadly I lack a soldering iron , soldering skills and I have sold or given away my recording devices, also I will happily let someone else take the abuse such as was heaped on my head when I did my cable tests a while back,  but we must have some hobbyists here with the means ?
 

 

[upstateguy]

Quote:

Sadly I lack a soldering iron , soldering skills and I have sold or given away my recording devices, also I will happily let someone else take the abuse such as was heaped on my head when I did my cable tests a while back,  but we must have some hobbyists here with the means ?
 

 
Why not just run some diffMaker files?  Even a false positive would be welcome at this point.
 
A lot of pontificating going on, the same as the cable anecdotes, but no one is willing to offer up any proof.
 
I suppose that until there is some sort proof, the premise that "all dacs, amps, and opamps sound the same" remains totally unfounded, for those of us in the back of the bus.
 
 

 

[Willakan]

Okay, nobody is declaring that "all amps, dacs and opamps sound the same." That is a fundamental misrepresentation of those people who hold superficially similar beliefs. For example, I do not believe that "all amps sound the same", but all well measuring solid state amps, operated within their limits, do.
 
To clarify my position on opamps: Modern, well-designed opamps, aimed at audio, implemented in a circuit that takes account of their voltage requirements, impedance and the rest, will sound the same.
That sounds like a lot of caveats, but they are generally met.
 
I disagree strongly with your statement that such beliefs are unfounded. I gave an earlier example, stating that if two things measure identically as applicable to audibility (for example, a THD reading of 0.0005% is effectively identical to a reading of 0.0001% under this condition), that they should sound the same is firmly established. If someone says they sound different, it is their job to prove that. Asking for null tests is all very well, but not many people have tonnes of measuring equipment lying around. If two things measure the same, why exactly should they not sound the same?
 
With opamps, for example, it would be even more complex - you can't just shove them all into a CMOY and measure bigger-than-expected differences (for aforementioned reasons).
 
However, to reinforce this viewpoint I felt it neccessary to have a look at two opamps which supposedly sound considerably different - the OPA627 and the AD8610. Let's compare measurements.
 
THD+N: This is actually compared in the datasheets (the AD8610 boasting how it outperforms the OPA627 - especially above 20khz). They are both in the region of 0.000x% - totally inaudible.
Noise: Obviously, for opamps this isn't just a signal-to-noise ratio. One is 6nV/(sqrt)Hz, the other is 4.5. Again, this is very low noise - the noise from the rest of the circuit will render this insignificant.
Dynamic Range: Not quite as easy to break down into two numbers, but both opamps are considerably more than fine in this regard.
Slew Rate: 50V/ms vs 55V/ms. Both plenty enough for any audio application I can conceive.
 
Yes, there are other potentially relevant numbers, but those are the main ones (frequency response isn't even worth worrying about). I double checked this to make sure I wasn't embarassing myself with an article I found from an opamp engineer at Analog Devices. The article is pretty old (1998) but still highly applicable.
 
 
 
 
 
[size=xx-small][size=xx-small] [/size][/size]

 

[upstateguy]

 
Quote:

Okay, nobody is declaring that "all amps, dacs and opamps sound the same." That is a fundamental misrepresentation of those people who hold superficially similar beliefs. For example, I do not believe that "all amps sound the same", but all well measuring solid state amps, operated within their limits, do.
 
To clarify my position on opamps: Modern, well-designed opamps, aimed at audio, implemented in a circuit that takes account of their voltage requirements, impedance and the rest, will sound the same.
That sounds like a lot of caveats, but they are generally met.
 
I disagree strongly with your statement that such beliefs are unfounded. I gave an earlier example, stating that if two things measure identically as applicable to audibility (for example, a THD reading of 0.0005% is effectively identical to a reading of 0.0001% under this condition), that they should sound the same is firmly established. If someone says they sound different, it is their job to prove that. Asking for null tests is all very well, but not many people have tonnes of measuring equipment lying around. If two things measure the same, why exactly should they not sound the same?
 
With opamps, for example, it would be even more complex - you can't just shove them all into a CMOY and measure bigger-than-expected differences (for aforementioned reasons).
 
However, to reinforce this viewpoint I felt it neccessary to have a look at two opamps which supposedly sound considerably different - the OPA627 and the AD8610. Let's compare measurements.
 
THD+N: This is actually compared in the datasheets (the AD8610 boasting how it outperforms the OPA627 - especially above 20khz). They are both in the region of 0.000x% - totally inaudible.
Noise: Obviously, for opamps this isn't just a signal-to-noise ratio. One is 6nV/(sqrt)Hz, the other is 4.5. Again, this is very low noise - the noise from the rest of the circuit will render this insignificant.
Dynamic Range: Not quite as easy to break down into two numbers, but both opamps are considerably more than fine in this regard.
Slew Rate: 50V/ms vs 55V/ms. Both plenty enough for any audio application I can conceive.
 
Yes, there are other potentially relevant numbers, but those are the main ones (frequency response isn't even worth worrying about). I double checked this to make sure I wasn't embarassing myself with an article I found from an opamp engineer at Analog Devices. The article is pretty old (1998) but still highly applicable.
 
 
[size=xx-small][size=xx-small] [/size][/size]

 
 
I do not believe that "all amps sound the same", but all well measuring solid state amps, operated within their limits, do.
 
According to your "viewpoint",  my Hornet, Portaphile, M^3 and GS-1 all sound the same. 
 
You may disagree strongly with my statement that such beliefs are unfounded, but until you are able to offer, at least, some rudimentary diffMaker proof, it remains nothing more than "your" viewpoint.
 
I'm also not sure that looking at opamp measurements, while ignoring components and implementation, is a valid way of proving that all SS amps sound the same.
 
As a final thought for this post, Bob Carver was able to win the "Challenge" by altering the sound of his SS amp to match that of a Mark Levinson tube amp*.  This in itself demonstrates that SS amps can sound different from each other.  And IIRC, Carver went on to market and sell "t" versions versions of his SS amplifiers which duplicated the sound of both Mark Levinson and Conrad Johnson designs. 
 
So, all well measuring SS amps, operated within their limits do not sound the same, do they? 

 
*Edit:  The amplifier Carver duplicated was the Conrad-Johnson Premier Five, not a Mark Levinson.