[bigshot]

perhaps it's the weakest link in the chain that prevents some people from hearing the differences various components make.  Maybe an amp that isn't resolving enough to hear the differences between dacs, or a dac that isn't delivering a high enough quality signal to be able to demonstrate differences between amps, or some other factor like a usb transport or an on-board computer audio chip?

The weakest link in the chain is *ALWAYS* the headphones or speakers. Electronic components that are operating to specifications are so far below the range of variability of transducers, the quality of the electronics is moot. The specifications of amps are measured directly without a transducer in the chain. That is the only way to quantify the minute differences between them. But we don't listen to DACs or amps plugged directly into our brains. We listen through our ears using speakers and headphones. And the level of variability of speakers and headphones totally swamps the minute differences in electronics.

 

[upstateguy]

 
Quote:

The reason I kept NOS DACs out of it is that many NOS DACs measure horribly - I would expect this to result in oft-audible differences. ....<snip> I'm not a NOS fan

.
 
Regarding DAC chips, they can be implemented in different ways - the aformentioned NOS approach, for example. However, the overwhelming majority of DACs are implemented in a way that does much the same thing as other DACs. I might have multiple chips, one per channel, to improve SNR or I can resample to various different frequencies. However, these differences in approach generally result in small differences which are unlikely to be audible - with the exception of the minority approach such as eliminating oversampling. So I can say the majority of vaguely modern DAC chips, implemented in vaguely conventional ways, sound the same. My original statement was perhaps a little too sweeping, but not by that much.
 
Opamps is a bit of a horny one for me personally. Tangent's opamp sound signature analysis has several aspects of it which makes it inapplicable to opamps as used in DACs:
1. The cmoy is a design which drives a difficult load (a headphone) directly with an opamp. This means that factors that would normally be dealt with with MOSFET buffers and the like in better amps can affect the opamp itself - for example, many cheap cmoys have uneven frequency responses.
2. Opamps, even in the simplest of circuits, are not neccessarily drag-n-drop. The "opamp rolling" which many people participate in always makes me feel slightly uneasy, because dropping in one opamp which has vaguely similar specs and hoping that the circuit won't mind is not neccessarily a good idea. However, I find it extremely unlikely that Tangent didn't take this into account - indeed, he speaks of oscillation and the like in his article.
3. In the end, it was a sighted test, whilst it is also unclear whether the volumes were matched. Anyone who has done blind testing knows how monstrously obvious differences can suddenly disappear.
 
However, in the end we are not looking at the measurements of the opamp - we are looking at the measurements of the whole. If the designer has used some weird and wonderful opamp and still managed to turn out a measurably exemplary product, bully for him!
 
As to your closing points, no-one is suggesting that every DAC and amp under the sun sounds exactly the same. The title of the other thread was more a provocative point to get people posting. However, there is no reason why a Benchmark should sound different to a DACmagic, nor that to the $6000 dollar dcs Puccini - as they all measure very well. Hell, you could probably go considerably cheaper than the DACmagic. If your point was that not every conceivable DAC and amp will sound the same, consider yourself vindicated

 
Regarding differences as expected/not expected, it's difficult to draw anything from this as to their nature. Perhaps you were subconciously hoping for this cheaper DAC to turn out better than it's peers. Alternatively, there were real differences that would also be easily measurable as significant - it seems unlikely, but not having the measurements of any of those products in hand, I cannot instantly dismiss it. However, were there significant differences in the measurements from what was expected (0.5% THD for example) then I could safely say that rather than one of the DACs being special, the others were merely inferior to what you would expect. Again, it's difficult to draw any conclusions from your specific example.

 
First:  You haven't addressed the PS issue.  Is a PS able to affect the sound of a device?
 
Now let's see what we've established:
 

1. All dacs do not sound the same since there are as many non-over-sampling dacs around as over-sampling dacs.

1. We can't even say that most dacs sound the same because we haven't established the effect of things like analog sections, opamps, mosfets, board topologies and PS.

1. At best we can say that some dacs might sound the same because they use similar topologies, similar components, similar chips and similar PS . 

 
Regarding dac chips:
 
You have said, "So I can say the majority of vaguely modern DAC chips, implemented in vaguely conventional ways, sound the same."  I don't know if that's true of not, but logic tells me that broad statements like that are probably not true or there wouldn't be very many dac chips around and there would be little incentive to develop new ones. 
 
You have not addressed the newest ESS 32 bit chip.  Does this chip have more resolution than any of the other vaguely modern dac chips you mention? 
 
Regarding opamps:
 
It seems to me that you have danced around a direct answer to the issue of whether or not different opamps can produce different sounds the way Tangent explains they can and then you introduce mosfets into the opamp discussion.  
 
Assuming that "designers" use various yet appropriate opamps and mosfets, let me ask you:
 

1. Can using different mosfets result in different end product sounds?
2. Can using different opamps result in different end product sounds?
3. Do opamps sound like mosfets?

 
You also make the claim that there is no reason why a Benchmark should sound different to a DACmagic.   Is that a supportable clalim?
 
USG

 

[mrspeakers]

Quote:

 
The concept of them sounding the same is not reliant upon null testing - it simply is based on the measurements. It is true that there is no set boundary where we say "Yes! This is a good solid state amp" (leaving tube amps and designs that alter the sound out of it, as they are designed to do so). However, when every measurement is vastly beyond the threshold of audibility, it is not unreasonable to say it is a well engineered amp. Take the hypothetical amp I now place before you - I have chosen measurements which are commonplace for good solid state amps:
 
...
 
Hence I feel relatively confident, when faced with sets of excellent measurements, to conclude that the systems from which they are taken produce sound that is effectively identical. The reason null tests do not abound is that the sort of people with the equipment and expertise to perform them likely feel satisfied that they are not required.

 
I am not a professional amp designer, just a hobbyist EE who does sometimes do high-end speakers.  I have to say there is a lot of interesting information in this discussion, and hopefully it will avoid going off the rails like the silly DAC thread did, so thanks everyone for such interesting reading.
 
Willaken, while much of what you write is interesting you make a fundamental assumption in "measuring" which is not correlated to real-world performance of systems: load.  
 
Virtually all "measurements" of amps rely on specific, usually totally resistive loads.  Headphones and speakers often have non-linear impedance, and this couples with factors like the output impedance of an amp to actually create variations in power-spectrum of up to 3dB in variance.  That's a factor of two in power.  In other words, unless you are driving an ortho with a pretty ruler-flat impedance curve, just the nature of the headphone interaction with a given amp will absolutely change the sound.  Some people can pretty easily hear 1dB in bass boost, others can not.  Now if you level match at 1KHz, one system will be perceived as having slightly more impactful or extended bass, even if the amps test identically into a resistive test load.
 
There are other factors related to feedback, transient intermodulation, and a few other factors that vary as a function of load.  
 
Also, I have clearly heard in my amps that the less feedback I used globally, the more open and relaxed the sound becomes, but it's NOT from distortion, which I can usually pretty easily hear.  Even on the fanciest test gear the lower feedback or no feedback version sounded better but tested worse.
 
Another example was when I was designing some speakers and my ex wife had been listening because she was curious.  Later that day I swapped out a single quality polystyrene cap with a cluster of parallel caps from a crazy expensive source.  She walked in and said "What did you change?  The treble sounds so much better..."  I had been unable to measure any difference, but was thinking I heard something, and was worried about placebo clouding my perception because I couldn't easily A/B.
 
Another interesting example is the cult of single-transistor amps in Japan.  Someone once took me to a "show" where these guys were swapping the transitor in their circuits out and trying different ones.  Some circuits had feedback, others didn't, but you could hear the difference with some of the transistors pretty easily.  Nobody was doing any measurements, though...
 
So to sum up my point, the challenge with testing is it can only show certain results under certain conditions and it is possible we just don't know what to measure that correlates to perception, vs to some simple metric like THD.  Measurements are the beginning but not the end of understanding.  
 
As a result, when I do do designs, I test then listen, and sometimes blind A/B.  Sometimes it's been clear I could hear something different but couldn't measure it, and other times I could not hear one.  BTW, we did use null testing too, including analog when we wanted to try to measure against ...
  
 
 

 

[bigshot]

Headphones and speakers often have non-linear impedance, and this couples with factors like the output impedance of an amp to actually create variations in power-spectrum of up to 3dB in variance.

That isn't important because speakers are never anywhere close to +/- 3dB anyway. Even if a magically flat set of speakers could be designed, they would change when you put them in a room. That's why all speaker systems that a listener wants to get anywhere close to flat require equalization. In general, you keep things up to the transducers balanced so different sources don't sound different, and all the adjustment comes at the final stage.

I don't use headphones much, but I imagine you would match the cans and the amp to minimize problems like that.

 

[mrspeakers]

Quote:

That isn't important because speakers are never anywhere close to +/- 3dB anyway. Even if a magically flat set of speakers could be designed, they would change when you put them in a room. That's why all speaker systems that a listener wants to get anywhere close to flat require equalization. In general, you keep things up to the transducers balanced so different sources don't sound different, and all the adjustment comes at the final stage.

I don't use headphones much, but I imagine you would match the cans and the amp to minimize problems like that.

 
Not important?  That is your opinion not a "fact," and frankly it's wrong because you missed my real point or just didn't understand it (granted, it could be me being a bad writer or because this IS a complex topic, but it does annoy me that you are being dismissive and absolutist).  
 
To be clear: 
 
Amp/speaker/phone interaction creates a relative change in performance that is specific to EACH output device.  Essentially, an amp/headphone is a voltage divider, and the power delivered to the load as a function of frequency will vary if the load impedance is not flat.  It is real and easily measured, and this alone will make my JH16 sound different as I switch between amps, because it does have a low impedance but the curve is not flat at all.  I like the sound much more on low output impedance amps, it not only sounds flatter, it IS flatter.
 
In my experience, the effect is actually much greater for headphones than speakers because so many 'phone amps have output impedance of 10ohms or higher, and the higher the output Z relative to the load, the greater the effect on frequency is if the phone does not have a ruler-flat impedance (like an LCD-2).  This alone will absolutely change the timbre of any device with a complex impedance.  
 
Why is this different from the +/- dB of a speaker?  That is a measurement of the headphone or speakers' frequency response in a given test environment.  If you change the amp in a test setup from a low impedance relative to the load to a higher relative impedance you will see the frequency response shift to a warmer shape, and the less flat the impedance of the phone is, the greater the change will be.
 
Now, can everyone hear a 1-3dB change in tonal balance through a broad range of spectrum?  In my experience in blind A/B in designs, yes they can.  If it's not A/B, the change is more subtle and some will notice, but not all.  
 
I had my DACPort redone with a 1ohm impedance and definitely hear a change in my JH16, but not in my T50.  Why?  Most likely because the T50 impedance is flatter and it's also a bit higher relative to the output impedance.
 
 

 

[maverickronin]

@USG
 
I think most of this problem is coming from differing definitions of what we mean by "sounds the same".
 
If you mean it all has to measure completely and 100% identically, then no nothing "sounds the same".  If your test gear has sufficient resolution then not even 2 products that came off the assembly line one after the other will measure exactly the same.
 
If the question is instead whether anyone can tell them apart with only their ears then a lot more stuff will "sound the same" despite the differing implementations and topologies.  Once the imperfections introduced by the equipment are below your ability to perceive them it doesn't matter if they beat your senses by 1dB or 100dB because you'll never know without something more sensitive than your ears.
 
All of the things you mention (opamps, DAC chips, power supplies, etc) will have an impact on the sound but what matters is how they're all put together.  They all interact with each other and the final result is what's important.  Isolating one component and trying to infer the final sound from it alone is almost as silly as saying that you know you won't like chocolate cake because you know you hate the raw eggs in the batter.

 

[Albedo]

People that try out some different op-amps to get closer to their liking -> http://www.diyaudio.com/forums/class-d/88117-ncd-questions-30.html#post1106028
 
..wish it was simple, that it all was a wire with a gain from DAC to transducer.

 

[upstateguy]

Quote:

@USG
 
I think most of this problem is coming from differing definitions of what we mean by "sounds the same".
 
If you mean it all has to measure completely and 100% identically, then no nothing "sounds the same".  If your test gear has sufficient resolution then not even 2 products that came off the assembly line one after the other will measure exactly the same.
 
If the question is instead whether anyone can tell them apart with only their ears then a lot more stuff will "sound the same" despite the differing implementations and topologies.  Once the imperfections introduced by the equipment are below your ability to perceive them it doesn't matter if they beat your senses by 1dB or 100dB because you'll never know without something more sensitive than your ears.
 
All of the things you mention (opamps, DAC chips, power supplies, etc) will have an impact on the sound but what matters is how they're all put together.  They all interact with each other and the final result is what's important.  Isolating one component and trying to infer the final sound from it alone is almost as silly as saying that you know you won't like chocolate cake because you know you hate the raw eggs in the batter.

 
 
I think most of this problem is coming from differing definitions of what we mean by "sounds the same".
 
I don't think that's the problem Mav.  In it's most basic form there's "sounds the same", "sounds similar" and "sounds different" and the party line has been "sounds the same".  I'm not so sure any more.
 
If the question is instead whether anyone can tell them apart with only their ears then a lot more stuff will "sound the same" despite the differing implementations and topologies.
 
Last week I would have said a lot of stuff will "sound the same"  but now I'm questioning this hypothesis because I'm starting to feel that, if for instance, one dac is using a particular opamp in its analog circuit, and another dac with the same topology and dac chip, is using a different opamp or a mosfet, there is a good chance that these two dacs are not going to sound the same. 
 
I suspose that its possible that some opamps and/or mosfets will sound similar but it's also possible that many don't and if the devices under review have dissimilar sounding opamps or mosfets they will not sound the same.
 
To date, no one has been bold enough to step forward and say all opamps and transistors sound the same, because they would have to offer the same type of proof we require of the cable believers.  I've issued several challenges for diffMaker files and desspite the possibility of false positives there have been no takers so far.
 
All of the things you mention (opamps, DAC chips, power supplies, etc) will have an impact on the sound but what matters is how they're all put together.  They all interact with each other and the final result is what's important.
 
It is my belief that opamps, mosfets, PS, topology and shielding all affect the final sound to the extent that if you listen to two devices with different complements of components, they will sound different from each other, in a greater or lesser way.  If I had to guess on the order of the things I've listed, I'd say that opamps and mosfets would make the biggest difference followed by PS.
 
It also hasn't been established if all dac chips sound the same.  And I'm beginning to think that there is more resolution to be gained from the new 32 bit Sabre chips, so that would be another form of not sounding the same.....
 
If you reduce it to audible differences in wire, I'm all in for no differences, but I'm not getting it for complex devices.

 

[upstateguy]

Quote:

People that try out some different op-amps to get closer to their liking -> http://www.diyaudio.com/forums/class-d/88117-ncd-questions-30.html#post1106028
 
..wish it was simple, that it all was a wire with a gain from DAC to transducer.

Ah, what do they know anyway.

 
Hey, where is 63°N, somewhere in Sweeden or are you referring to Triakel?
 

 

[Albedo]

Gain on a wire, that's my dream... in some way I'm a bit envious.
 
About 63°25′N, 10°23′E then..  dubbed Norway's technological capitol, mainly because of SINTEF.. the largest independent research organisation in Scandinavia.

 

[bigshot]

Not important?  That is your opinion not a "fact," and frankly it's wrong because you missed my real point or just didn't understand it

I think the misunderstanding is that you are talking about headphones and I am talking about speakers. You might want to use headphones with several different amps, but once you've amped your speakers and EQed them to get as flat as you can, there wouldn't be much reason to swap in a different amp.

 

[bigshot]

Once the imperfections introduced by the equipment are below your ability to perceive them it doesn't matter if they beat your senses by 1dB or 100dB because you'll never know without something more sensitive than your ears.

It also doesn't matter if the imperfections are below the amount of error of your speakers or headphones. What's the point of lowering a noise floor to -100dB if your living room has a noise floor of -50dB? +/- .1dB in your frequency response doesn't mean much if your speakers are EQed more like +/- 2dB. The transducers are always the final arbiter of the sound because they're making the sound and they never perform as good as digital sources do. You can't listen to music without transducers, so splitting the atom and achieving spectacularly good specs beyond the range even bats can hear isn't going to sound any better than your headphones or speakers can reproduce it.

People talk about better transducers being more revealing of imperfections, but they usually mention that in reference to anomalies that are far beyond the scale of what the best transducers can even reproduce. It's all a matter of scale.

 

[upstateguy]

Quote:

I think the misunderstanding is that you are talking about headphones and I am talking about speakers. You might want to use headphones with several different amps, but once you've amped your speakers and EQed them to get as flat as you can, there wouldn't be much reason to swap in a different amp.

there wouldn't be much reason to swap in a different amp.    Unless you find an amp that has more resolution......
 
But I'm starting to think that headphones have it all over speakers when it comes to being able to hear increased resolution.  I've listened to speakers as long as you have, but when I really want to hear what's going on in a recording, I always go to headphones.
 

 

[bigshot]

The resolution of any reasonably good amp is far beyond what flapping paper in big wooden boxes can reproduce. The way to improve the resolution of the flapping paper is to equalize so spikes aren't causing problems with the way the paper flaps or masking the octave above.

Headphones are great for listening to surface noise on records and hiss on tapes. I use them myself for noise reduction because they're so unforgiving. Cans bring everything close up like a magnifying glass. It's all about detail.

Speakers are better for putting sound all around you and creating bass you can feel in your gut. The reflections off the walls make it sound like the music is right there in your house with you. It's much more present, balanced and lifelike.

Headphones are better for examining music and speakers are better for experiencing it. It's the difference between resolution and presentation.

 

[upstateguy]

 
Quote:

The resolution of any reasonably good amp is far beyond what flapping paper in big wooden boxes can reproduce. The way to improve the resolution of the flapping paper is to equalize so spikes aren't causing problems with the way the paper flaps or masking the octave above.

Headphones are great for listening to surface noise on records and hiss on tapes. I use them myself for noise reduction because they're so unforgiving. Cans bring everything close up like a magnifying glass. It's all about detail.

Speakers are better for putting sound all around you and creating bass you can feel in your gut. The reflections off the walls make it sound like the music is right there in your house with you. It's much more present, balanced and lifelike.

Headphones are better for examining music and speakers are better for experiencing it. It's the difference between resolution and presentation.

+1  .......I agree with you completely.