But the bit about technical performance isn’t correct.
I once held the view that there was something else not captured by the graphs, and that’s technically still true, but the mistake is thinking this is something other than FR.
Sorry I may be slow, but isn’t there a lot of technicalities independent of FR that “higher-tier” headphones excel at vs “lower-tier” headphones?
Examples I’ve found: (including both stock sound and harman target EQ , though I do admit harman target EQ among headphones isn’t equiavalent)
Nope. That’s just FR relationships. ‘Technicalities’ is a common misconception, and one the audio community has long assumed to be true - as I once did, but the better we understand headphone performance in situ, the more we realize it’s all just FR.
It doesn’t diminish your experience of those effects, but the cause is still FR at the eardrum.
C’mon, @Resolve. What about things are aren’t technicalities but gross factors in design. Like larger drivers that move more air?
Are you talking only open backs, or are you counting echos and internal reflections of closed backs as simply aberrant FR?
You saying that things like physical distance of drivers to ear boils down to FR?
On that last, I point to LCDi4 vs almost any other IEM. Not saying that FR isn’t the primary component. but as far as I know a working brain isn’t independent of a beating heart.
Also something you’d see in FR, and that’s because of headphones being minimum phase. So all those time-based metrics are just a different view of FR, and if not you’d see that in excess GD plots. Still a chance there’s something there, but the point is it’s unlikely to be outside of FR.
This actually has a massive impact on FR. The one consideration here is perceptual psychoacoustic aspects to the experience, where things like driver distance, angle, openness that sort of thing are influential. The door is very much open to that stuff being perceptually relevant.
Having multiple headphones allows for matching the various genre of music to match up with a headphone that best fits. The Sennheiser HD 600/650/660 S2 works well with a lot of jazz and classical. It’s fun to change things up occasionally to achieve a slightly different playback between the three.
Sometimes a closed back headphone is required. Finding one that works well is a challenge, to be sure. I have found a few that are acceptable.
The Verum 2 is a great all around headphone for all types of music. This headphone is one that does not need EQ (for me) to sound right. One could spend at LOT more and not achieve improvement.
Lastly, if one has more than one headphone amp, having headphones that match the power level required and impedance for the headphone helps. The Sennheiser headphones sound great with OTL Tube amps and most SS amps. The Verum 2, at 12 ohms impedance, is not optimum for the OTL amp. Some SS amps are not optimized to run the very low impedances of the planar headphones.
In other words, the door is very much open to 100% of the stuff that actually matters to humans…
When a person doesn’t have a flashlight they might hunt for their lost car keys under a street lamp…even after losing the keys 100 meters away. There’s no way to reduce the relevant factors to FR and have a meaningful discussion of human perception…
On the contrary, we know FR is the most relevant factor, even in a predictive manner for humans. It’s entirely unclear to what degree things like openness and driver angle matter, independent of any FR changes in situ at the eardrum. That’s why it’s currently being researched.
It’s not about confidence, it’s about evidence. We have all kinds of evidence about the relevance of FR. Every time we go looking for evidence of acoustic properties outside of FR like ‘technicalities’, we end up being able to explain it with FR in situ. But… suppose we do find evidence of something else - and there’s every reason we could, I expect things like openness could end up being relevant - then we have something to talk about.
It’s also not as if focusing on the part that can be correlated and explained with FR entails a 100% refutation of the things that haven’t yet had similar explanations/correlations proven: As is always said, we leave the door open to things like acoustic openness, comfort, and other non-sound-related psychoacoustics factors to have meaningful effect.
However, we also don’t just throw our hands in the air ignorantly claiming complete irrelevance of FR to things it can contribute to or fully explain, like most of the things people call “technicalities”. Doing so is simply the result of actually engaging with frequency response in earnest instead of futzing with EQ for 10 minutes, being unhappy with the result, and deciding that said testing is sufficient to conclude that FR ruins technicalities (or seeing a graph, misunderstanding it, and deciding that means FR can’t explain technicalities).
The vast majority of the time my interactions on this topic are the result of people thinking FR = the graph, and assuming that this is all there is to that headphone’s measured performance.
We’ve been particularly charitable to the notion that there could be something else other than FR that is perceptually relevant, largely because it’s good practice to be open to new evidence. But given how much of a gap there is in terms of understanding what the FR actually is in-situ for people, it seems insane to me that we’d jump any conclusion that resembles “we should ignore FR entirely because there could be something else”. That’s what I see overwhelmingly occur in audiophile spaces, and it’s rather unfortunate.
I’m certain that physical distance has a massive impact on FR.
I still take issue with closed back reflections being FR. More of a time domain issue.
If you look back at my review of the Audeze LCD-2 closed years ago, I tried to describe the sound of a classical guitar in a live stone church as heard on the the headphones. I could close my eyes and distinctly hear both the large room of the Spanish church and a close shell reflecting. Truly weird.
I didn’t hear the shell on any open backs I tried. Rosson, Hifman HE-560.
I just can’t put that experience into the FR bucket.
I think once I appreciated that “FR” (and the role claimed for it) does not mean “the graph,” the conversation becomes interesting rather than something that needs to be argumentative.
And if by “FR” we mean “the properties of sound waves,” then to my mind it’s not even debatable anymore. Sound waves are characterized by frequency, amplitude, and phase. And that’s it. There isn’t anything else. No matter what goes into the driver and anything around it, it’s only going to produce waves with those properties.
Now, as to how close we can get to measuring those things . . . that’s where things get quite interesting. I suspect we’re pretty far off.
It should get interesting. But of course more commonly this idea gets met with resistance, perhaps because people don’t like the idea of their experiences being ‘reduced’ to a single metric like that. Okay, I get that. But at the same time… that should be fascinating to people. In no way does the fact that we can understand this stuff better with an existing metric diminish the experiences people have that are contrary to the graph, if anything in situ FR is precisely the answer to why there’s more to the experience than the graph.
Not being able to put it in the FR bucket seems strange considering the considerable difference in frequency response between LCD2C and the others. The LCD 2 Closed is almost certainly emphasizing a lot more upper midrange and treble, thus bringing forth reverb cues that otherwise wouldn’t be so forward in the mix.
In other words, its not the headphone itself having time-variant behavior, its the frequency response of the headphone uncovering/highlighting time variant behavior (like reverb/delay) in the mix you’re listening to.
Indeed, frequency response is simply the pattern, magnitude, speed, and phase of the oscillation of air (and the eardrum/microphone membrane). People would do well to understand its completely separate from the flawed measurements (and flawed interpretations thereof) you’ll often see online.
Defining what the Frequency Response actually is where the debate lies. Whilst there are sets of measurements used to design and evaluate frequency response curves for headphones, there remains debates about which tools are the most accurate.
Moreover, there remains lively debate about the accuracy of each of the tools available. Then, take into account the fact that each person has their own unique hearing (hearing, canal, etc.). The graphs do provide good information, but they are not entirely conclusive. How many times have people read that headphone X and headphone y measure almost the same, yet sound different, or a graph shows a deviation at a specific frequency, yet the reviewer does not hear any impact with the frequency deviation.
IMHO, I think distortion is more audible than what is claimed by many. I’ve read that techs who put speakers together have heard frequency deviations and distortions at much less than 1%. Also believe there are phase shifts that occur when EQ is applied that can adversely impact sound playback. The Sonarworks app also shows that there are phase shifts and latency that occurs when EQ is applied. Perhaps that’s why more often than not, EQ is a lot like closed headphones. May be necesscary, but not easy to get right.