I would probably be cautious of using equal loudness work as a basis for estimation of HRTF differences - equal loudness studies have bounced back and forth on the correct levels particularly at low frequencies over the years due to methodological variations.
Comparatively, the aforementioned work of Hammershøi and Møller shows little low-frequency HRTF variation, and of course Olive, Welti, & McMullin 2014 found marginal effect of culture on preferences.
Large scale norming research has been in use for 100+ years. It is routine in education, job placement, and medical research. Yes, human anatomy and physiology are relatively consistent whereby you don’t need a ton of people to understand the bulk of all humans and can indeed reliably predict many things. However, there are exceptions, there are variant subpopulations, and there are endless methodological and political controversies when one “goes there.” […mumble…watch or read the news headlines of 2020 and 2021…]
Norming research tends to be resource/cost intensive, so it’s driven by well-funded organizations and businesses with direct financial and operational needs. The US military has used normed IQ and aptitude testing for recruit and draftee placement since before WW1. Over the decades IQ testing in particular came to be criticized for numerous reasons, whereby these types of tests have often been replaced by achievement tests. Achievement and learning can strongly correlate with the deterministic/absolute IQ framing of data, but account for individual change. Whatever is used today has been reviewed and modified many times. This type of testing spread widely into education and career tracking for small children generally.
The take-away is that norming data you find will be largely driven by convenience samples (i.e., college students required to participate) or by an industry sponsor (i.e., testing only those buying products or employed in this industry). There will be gaps, and they may miss the training factor entirely. Perceived audio quality is not the same thing as mechanical hearing potential. Hearing potential is a limiting factor, but the final perception is akin to an achievement test. Test-rig measurement data and normed perceived preferences may indeed be highly correlated, but they are absolutely not the same things.
“There are many ways to skin a cat.”
“There are 50 ways to leave your lover.” - Paul Simon
I didn’t mean to day that I felt that this was indicative of any sort of rigorous proof - only that it was the only significant curve variation that I found. Also whether it’s accurate or not I don’t know, but the Japanese audio industry seems to use it, so my speculation was that it might at some level explain some amount of bias that goes into Japan’s productions.
My findings and believe both in practice and empirically are that we all hear much more similarly than different, and to a pretty high degree.
This is also what I believe. The brain is quite amazing at normalization.
2 posts were merged into an existing topic: DSP, EQ and other Plug-Ins
It seems this thread has been inactive for a bit, so here’s my attempt to wake it up. 
I see/hear the phrase “X headphone is more detailed than Y headphone” a lot, which I think is an interesting sentiment. Objectively, if the detail is all captured in the recording, can one headphone truly be more detailed than another? Is the headphone actually responsible for detail?
I take this as the potential speed of the driver: does it start and stop fast enough to keep up with complex sources or does it lag behind and thereby blur the sound? Starting and stopping can involve physical movement (dynamic drivers) or direct sound output (planars).
If the speed of the system (DAC + amp + driver) exceeds the potential of the source then they can sound etched or edgy. If not as fast they’ll sound fuzzy, loose, or noisy. I don’t tend to pick up on speed with slower, simpler music (e.g., vocals, acoustic guitar), but do on complex orchestral or studio pop (e.g., panning, overdubs, etc.).
The Clear is pretty fast and I hear sharp edges on the THX 789, but not most other amps.
The Clear is quite resolving and…well, clear, but is it actually providing “detail,” or is it simply providing more clarity so that you can hear the detail that’s already present in the recording?
For me, details standing out are hugely influenced by true timbre. Say I’m listening to a track I know, and a percussive strike on a wood block really grabs me. I say wow, never heard that before! So real!
But if I then go to to an inferior headphone, dac, whatever, and listen for that sound, yep it’s there. But it’s not real and alive, so I truly never noticed it at all.
A headphone can obscure detail that’s in the recording, whether in frequency domain or in time domain. An amp can fail to respond quickly enough to capture the most ephemeral minutiae in the time domain. A DAC could theoretically be insufficiently resolving or something like too warm to give treble sounds the amplitude they need. If all three playback devices are doing their jobs fully, then the recording would have to be the bottleneck – which takes us back to the microphone(s), capture medium, mixing, mastering, etc.
IOW, the medium is only the message when engineering considerations permit, grin.
This is a very interesting question, and in my view there are so many variables that muddy the waters. Take the Focal Utopia for example. That’s lauded by myself and many others as the most ‘detailed’ DD headphone out there - and as Tyll said it really allows you to ‘peer into the music’ better than others. Based on those accounts that would indicate you’re effectively looking throw a window that’s more clear, meaning that the details that exist in the music - that can be heard in other headphones as well - are simply more ‘clear’. But at the same time, this raises the question… why are they more clear? It may very well be the case that our attention is drawn to those details more significant as a result of certain characteristics of its frequency response. People often talk about how there’s a certain timbre as a result of the series of elevations in the mid treble on the Utopia, and perhaps that’s responsible for the same thing.
At the moment, in my view, the why of it simply isn’t obvious, and that means we have to not be satisfied with just target adherence. Is the experience of being able to hear those details more clearly better? I’d argue yes. So then does that throw a wrench into what’s desirable in frequency response if it is all just FR? Even when you run these ideas to their conclusion, I think we’re left without a satisfying answer, nevermind the fact that there are numerous counterexamples to each seemingly conclusive result.
Bottom line, we need more studies on this subject in particular.
Thanks. Yeah, it’s an interesting topic, one that certainly needs more studies and testing - not only regarding where detail actually comes from, but also what aspects of the headphone are critical in retrieving it properly.
I like the “looking through a window” analogy when describing a headphone’s resolving ability, or when speaking to how “clear” things sound, but I don’t’ think it applies to “detail,” since the detail is on the recording and the transducer is simply a means of retrieving it at varying levels. Some headphones are better at it than others, of course.
If I recall, one of the projects you’ve mentioned, that you cannot fully unveil yet, is focusing on detail retrieval and whether it can all be tied to FR or not. I’m looking forward to the results of that!
I think you guys are on to something hear regarding clarity vs detail. I have experienced this of late in regards to solid state amplifiers.
To begin with, my foray into extreme detail has been with one particular amplifier. This would be my most expensive stand alone solid state amplifier, the Ray Samuels HR-2. The design is antiquated mostly due to the AD797 opamp that at the time of the amplifier’s design in 2002 was at the forefront of performance, and it seemed many were seeking amplifiers capable of increased detail back then. In stock configuration, the HR-2 bests all of my new and old solid state amplifiers of my collection (the rest are $800 msrp or less) in terms of detail retrieval. I often refer to the Ray Samuels HR-2 amplifier as being my ultra-detailed amplifier. However, the performance in all other aspects are dry and cannot compete with any of my $400 and up modern amplifiers (all the usual suspects; Schiit Jotunheim2, RebelAmp, Singxer SA-1). It is defeated by them on many levels on all other characteristics unless one rolls modern opamps. So, as a result, I almost never mention this amplifier within my discussions.
Fast forward to today, to a modern day quality amplifier that integrates two decades old designs and is mostly an updated tweak on the decade and a half old AMB amplifier, the Rebel Audio RebelAmp. When listening to my Rebel Audio RebelAmp, the detail retrieval is maybe half of the Ray Samuels HR-2. However, with the combined performance increases of other superior characteristics such as an enveloping perceived sound stage and warm, smooth, rich, textured presentation, the RebelAmp somehow exudes clarity in spades despite the comparatively lower amount of detail. This has revealed to me my preference to desire clarity above detail. And this has demonstrated an amplifier’s ability to maintain engagement and enjoyment with reduced detail when other compounding factors of elevated performance are achieved.
My THX AAA 789 seems to have some sort of noise reduction circuitry that enables tremendous measurements and silent noise floor. However, this noise reduction is also introducing signal processing in a negative manner. Decay and reflections that would be present in a beautiful sounding amplifier, seem to be cut short with the THX AAA 789. Perhaps as Dudley_Doody has stated, “respond quickly enough” and ambience cues appear. However, I hear the THX AAA 789 jump in too quickly with it’s reduction circuitry as it kills any perceived ambience. For some, the resulting product is pleasing, coming across as an intimate presentation. However, to my ears is a very artificial presentation and my ears would rather achieve intimate presentations with a headphone that has this characteristic instead of through signal processing.
This same THX processing in my SMSL SP200 also yields a very narrowed sound stage.
I perceive this artificial clarity. THX seems to reduce spatial detail to recreate artificial clarity.
I’ve read various comments that THX/Topping technology involves “subtractive distortion.” At least perceptually, as I don’t have the equipment to test. In focusing on static charts and measurements, the engineers may have overlooked what’s needed for time-series playback quality, note coherence, and smooth transitions.
My benchmark test track for the THX 789 is Frank Sinatra/Carlos Jobim’s The Girl from Ipanema. I hear warm richness on anything else, but raw, rough edges and a too-hot microphone on the 789. This was recorded in the 1960s, so back in the day it was certainly perceived as thick warmth.
I think the 789’s noise reduction processing cannot distinguish between intentionally recorded air, breath, and room noise versus unwanted reproductive/playback noise. As such, it strips music bare and indeed causes artificial clarity…and thereby invented a new type of distortion.
@hottyson @generic Nope, you guys are wrong. The only thing that matters for amps in SINAD measurements. 
I was confused on clarity vs detail. I watched this video and after thinking on it and reading your post on the ¿What is detail retrieval? thread, I think I understand now. A headphone can only reveal the detail present in the recording. The headphone’s ability to do so is either the whole or part of the headphone’s clarity.
But it seems that many reviewers discuss them as two separate aspects of a headphone’s sound. 
Yes, precisely! The guy in the center on that video defines clarity really well - great find, BTW. The clearer a headphone is, the easier it is to hear the detail on the recording, whether the detail is true detail or artificial detail added by other things (amps, DACs, distortion, etc.). Even if it’s the latter, you can then focus on what’s causing “detail impersonation” and how to rid your system of it; and if you achieve that, you’ve actually further improved clarity, thus improving the ability to hear more true detail.
You’re right - reviewers do sometimes discuss the terms as if both are traits of the headphone, when really only “clarity” is. It’s really quite the interesting topic - one that I hope is further researched and discussed.
T1.1 is a very well balanced headphone. It was known as the jack of all trades when it came out compared to the HD800. The reason a lot of people have issues with treble with the T1 is because they don’t realize that the headphones jumps up to 1k Ohm impedance at lower frequencies. With modern amps, the OI is lower to accommodate for more headphones and to have a higher damping factor. Generally this means that a lot of amps aren’t able to provide the voltage that high impedance beyers need to produce a more balanced tonality which leaves the treble to be more emphasized than the rest of the range. Also, cheaper dac/amps have more glare than the higher end offerings which will make the situation even worse.
Long story short, T1 is a good headphone but does require the correct dac/amp (OTL tube amp) to go with it. The way amp manufacturers have been designing their amps means beyers get a really bad rep with high impedance drivers. That’s one of the reasons all the new beyers have lower impedance drivers now.
My T1 copy plays back very well with the ifi Diablo… very balanced across the FR…