But then that suggests that with respect to the auditory experience, there’s no there there. Is the suggestion that the test may have been poisoned to bias the listener into thinking there is no difference in the controlled situation? Or just misattribution of the difference. Because… I feel the real issue is how the community/industry deals with the controlled situation, where the differences disappear, not that they can suddenly hear new differences as a function of improper test methodology.
There may be no there there or a given test may be fatally flawed. Anything could hypothetically be true. The audio industry largely runs on hype and pumping new stuff to keep the cash coming in (as does every industry). Many tests are likely kept secret for competitive reasons – I heard that Bose is ultra secret from a colleage who interviewed there.
More generally, welcome to Theory of Mind and cognitive psychology. This is the branch of science that tackles epistemology and qualia. But, with tests of real people using machines in labs and hospitals–and a heavy reliance on interpreting probabilities using bell curves across many trials and many different people. Forced blind choices can work amazingly well at finding things people reliably perceive but DENY being able to perceive. Also see real-time fMRI brain imaging, where small components of brain processing “happen” but do not reach conscious self-awareness.
1 Like
Sounds like my Father who liked to quote, “Those who can do, those who can’t teach, and those who can’t teach teach teachers.”
Those who can test, Those who can’t test, test conditions, and those who can’t test conditions test tests. Note that I’m being very careful not to speculate about the smallest particle or quanta of testing, the testicle.
1 Like
Many thanks for the thoughtful reply, Luke. I am quite new to the analytical sides of this hobby.
In your experience, do you believe that a more 3-dimensional model of a DAC’s output would allow us to measure quantitatively some of the impressions that so many hobbiests have about the different tonal qualities of such instruments?
This branch of research grew out of (1) ship navigators who found that some people were faster or slower in recording light house sightings (i.e., setting the ship on different courses, (2) William James of Harvard U. converting a basement into a test lab for ancient philosophical questions – see Plato’s Allegory of the Cave, and (3) the military dealing with human differences in reading radar signals.
Regarding @Resolve and hallucinations, yes, people are often suspected of being ‘off’ or ‘odd’ or ‘wrong’ about their perceptions versus others. But then, research with data has typically found bell curves with human differences and performance. This applies to obvious stuff like height as well as the ability to see, hear, react, etc.
1 Like
I’d suggest you give the Schiit multi part story posted on head fi a read. The owner of Schiit does a lot of measurements and listening tests (level matched and blind sometimes). His journey, explanations and conclusions are much more nuanced than any summary.
But in case you’ve got no time or appetite for it, in short he believes that people should focus on getting the right headphones and speakers that suits their needs. Amps and dacs simply don’t sound very different from each other relative to the difference that exists in transducers. Some people do have the ability to tell amps and dacs apart, and a lot of people can’t, or don’t think the difference is large enough to care about.
He personally performs a huge variety of tests to ensure his products preform well, but it’s listening impressions that matters to him at the end. Not every test matters, as many products are audibly transparent, but people can and do have preference, and all normal listening at home are done sighted.
Ahh…that was you who posted that! I thought maybe it was just an advertisement. I’ll give it a good read tonight!!
One item I would like to bring up is variation in measurements from unit to unit for a given headphone. I have experienced this first hand with the Dan Clark E3 headphones. The first set I bought was recessed in the midrange enough to be quite noticable. I would up selling them, as I thought that was how they were voiced. I was wrong.
Several months later, decided to try them again, as I was still in thw market for a high end closed back headphone. The second set sounds outstanding. The midrage sounds “correct” with this set, quite different from the first set I owned. This set, I happily kept and use quite often.
So, my question is: How often does this happen? I would not have expected such a variation from one unit to another with high end headphones. Makes me wonder if this is one of the reasons why there are such diverse opinions about a given headphone.
1 Like
Gotta be pretty often. And pad wear changes things quite a bit too. My e3 sounds a bit closer to shouty in the mid range than not.
That’s why I’m not that keen on stuff like using other people’s eq profiles, even if they came off measurement rigs.
Good point. We all hear differently.
Case in point: One of my headphones that I like are the Ultrasone Edition 15 Veritas. I know a lot of people don’t care for them. However, using them with a Chord Hugo TT is a great sounding combination, the sum is better than the two parts. With other setups, they can sound quite sterile. With the Chord, they sound extremely good, which is why I have kept them. They also sound good with my other headphone amp, which is a Aurouasound HEADA amp.
The S-Logic for me works very well. For others, they hear no difference.
Whilst I agree the headphone transducer makes far and away the largest delta with listening, I’ve heard signifigant differences with both DAC’s and headphone amps. I have had the Benchmark DAC-2, and the RME-ADI-2. They are both fine examples of audio gear. However, I didn’t like them. They sound too sterile for my taste. For DAC’s I prefer the FPGA designs from Chord and PS Audio, They sound more natural to me.
The Chord headphone amp section in the Hugo DAC’s also sound good to me. My favorite headphone amp is the Aurouasound HEADA amp. Has all the best qualties of SS, whilst having a midrange and treble that almost would make one think it was a tube unit.
One thing that’s worth noting is that ‘FPGA DACs’ don’t technically exist. This is a bit of a misconception from some marketing put out by a few companies.
FPGAs themselves are just used either for digital processing or control of other components.
So as well as doing any oversampling/DSP etc, in the case of a dCS DAC the FPGA controls the ring-DAC.
In the case of a Holo Audio DAC the FPGA controls the R2R ladder.
In the case of a Chord DAC the FPGA controls the pulse array.
But many AKM/ESS DACs including both the ADI-2 and DAC-2 also use FPGAs too.
The actual DAC topology is a different thing, but some companies focused really heavily on the FPGA marketing and as a result there’s a bit of a misconception that they’re something separate.
It sort of reminds me of that scene from Mad Men:
Don: How do you make your cigarettes?
Rep: …Grow it, cut it, cure it, toast it…
Don: There you go! “It’s Toasted”
Rep: But everybody else’s tobacco is toasted?
Don: No, everybody else’s tobacco is poisonous, Lucky Strike is toasted.
Appreciate the feedback. Reminds me of the old saying, “Lies, Damn Lies, and Statistics”.
Recenntly auditioned a Topping DX9, and I was very surprised how good it sounded as a stand alone DAC. The built in headphone amp sounds OK, but the DAC section connected to a headphone amp (Aurouasound HEADA) sounds very good, IMHO.
Many people are in the camp that there is very little difference in performance with modern DACs. After auditioning the DX9, I have to admit there may be something to this.
1 Like
That’s my question – not whether SINAD tells the story, but whether anyone can tell two well-measuring DACs apart in a double blind study, because if not, the relative merits of SINAD are irrelevant and not worth debating. I’m not talking about the ability to distinguish between filters, as GoldenSound did in his test, just the DACs themselves with their settings optimized for transparencey. Because if no one can reliably tell them apart, then it is indeed game over for the entire DAC business – not to mention DAC reviewing business. Which would be GREAT for the average person, because then they could buy a DAC for a few hundred bucks and move on to transducers – forever.
2 Likes
Tangent, but I’m shocked they’re still around. They never come up in any discussions about IEMs yet they’re still able to sell four-figure products with poor tuning. Who the hell is buying these?
Oh so your algorithm showed you this clip as well, huh?
It depends on the DAC’s under test. A few weeks ago, obtained a Canever Zero Uno DAC-HPA. I was floored at just how good this unit sounds. Absolutely amazed. Prior to listening to this, had started to get pretty cynical about what differences could be heard between modern DACs. This unit changed my mind.
As Golden Sound stated earlier, there is a misconception about off the shelf chips and custom designs. The Zero Uno DAC-HPA uses a SARRE 9018, but it sounds NOTHING like the 9018 in say an Oppo Blu Ray player. Mario wrote his own code for filtering. The design differences with with DAC-HPA are too extensive to get into details here. For example, the output from the 9018 uses Lundal isolation transformers, there are six separate power supply transformers, the XMOS USB has a quasi battery isolation, the analog amps are ultra liner, with no caps or negative feedback. The output devices are individually matched, where only a small handful out of a batch of 1000 pass muster for use.
All these things make a difference.
This is easily the best headphone setup I’ve heard to date.
Hello, new to the forum. Thought I would jump into the deep end and just put my random thoughts of the last few day’s on here as a starter. (haven’t read anything in this thread or on the forum yet)
I have never been one for graph’s or charts, like at all. I have purchased… ~30? headphones and half of them have made the cut. I decided that my path in the headphone world would be one of learning by doing. I had no concept of what various terms meant in audio, rather I chose to purchase mid-fi product’s one at a time, each one an example of a specific trait or two. I have spent the past 5 or so years in the head space and while I could have one golden end-game pair of headphones… I am happy to have gone on a journey with those funds instead of taking a random shot in the dark. I have a “high end” pair of headphones on order currently.
The reason I’m in this forum is a video I watched the other day about being able to EQ virtually any headphone into any headphone and it got me wondering why I have rather expensive headphones on the way. Granted I don’t understand amp’s so have simply purchased some overpowered amps, but I feel like every last piece of gear I have is unique in some way. So why is that? My monkey smooth brain can only question things’ like cabinet resonance and such… If you make a pair of headphones out of cardboard with glorious drivers… aren’t they going to sound like crap with any amount of EQ? I think there is something wonderful about the mystery of audio, subjective construction across many avenues, power, dac, source, meat brain included makes this such a fascinating hobby.
But I also love the scientific approach that headphones is going after, even if it’s not my jam. If nothing else it’s a way for me to learn and experience different thing’s. I look forward to looking at people drawing light to various aspects of various gear and being able to go through my library again to see what takes I have on what. Hopefully my ole brain can keep up, cause that’s always a question.
1 Like
I think headphones generally are mysterious to most people. But a lot of it has been figured out, and continues to be figured out so long as we maintain a healthy curiosity about this stuff, and aren’t overly dogmatic about the things we learn.
There’s a tendency for people in this space to learn a little bit, and then fixate on just that bit, dismissing all the rest of the stuff that’s there to be learned/understood.
But… with the car analogy, I see it as if the driver is the engine. You can have the best engine in the world but it still won’t be great if the wheels are made of cheese. So it matters, but it’s what you do with it that matters the most. And the entirety of that system can be evaluated is in terms of sound propagation at the eardrum, which we can visualize as frequency response.
This is why it makes sense not to over-fixate on driver stories or special properties of various fancy materials. They may be genuinely doing the things manufacturer marketing copy tells you they’re doing, but that doesn’t mean it’s resulted in a good product. Moreover, a driver need not be expensive or have fancy materials to be good, to perform well, or have a good implementation.
Things that I look for in a good driver when building headphones:
- Low resonance frequency
- Low harmonic distortion
- Non-modal behavior (makes them easier to tune)
- Cost/weight
There are of course other things to consider, but for most modern drivers those things are easily achievable. So it’s really the above characteristics I’m focused on when it comes to driver performance.
2 Likes
Welcome aboard Klute.
People buy expensive headphones for all kinds of reasons. If you’re under the impression they’re only doing it for a certain frequency response, then you’re probably mistaken. (Gravely mistaken, I’d say.)
If you are looking for good sound quality though, then the recipe is probably the same regardless of how expensive the headphone is. You will get different opinions on exactly what that recipe is depending on who you talk to though. Resolve has given you part of his recipe. Mine goes something like this…
A very open, spacious, transparent sound with low distortion and low noise, good symmetry and imaging, good clarity and detail, good dynamic range, a good comfortable fit with good coupling, consistency, and pinna interaction, good extension in the bass and treble, and a well-balanced, neutral frequency response with no obvious unwanted resonances, colorations, timbral distortions, or dropped notes. I will probably think of more tomorrow. 
Overpowered amps aren’t always a good thing btw. For starters they use (and potentially waste) more energy. And if they aren’t well designed, and/or don’t have the appropriate gain or impedance for your headphones, then they can potentially add noise and distortion to your setup. Or cause some damage, if you aren’t careful with how you use them.
3 Likes
If it were true that by using EQ you could make bad headphones sound like good ones, then that would be that. Everybody would get $10 headphones from the bargain store, dial in an EQ setting, and everything would be fine.
First, some cheap drivers in the headphones don’t have the range of good ones, so you CAN’T just EQ.
Second, even if the frequency response is similar, a lot of other things won’t be. Here are a few:
- The amount of air the driver can move
- Internal reflections inside the headphone
- How changes in volume affect the sound
- Comfort
- Other design considerations - build quality - over, on, or in ear
4 Likes