I suppose there’s alot to think about on that, including the IP (intellectual property) implications.
Well, there’s a long story to do with all of it, but the gist of it is that we don’t want to use systems like squiglink because they’re prone to people making a lot of errors - insufficient guardrails for applying erroneous targets to systems that aren’t applicable to them. Moreover, the whole project of single line targets in the first place is somewhat of an antiquated idea, and so while I see the utility in having a system like that, it’s also baking in a concept that communities need emerge from.
But also yes, IP is definitely relevant.
1 Like
I understand some of your concerns on this. I guess you have to think about what you want your users to be able to do with the data.
One of the reasons I like the squiglink concept is that it’s pretty flexible. I can upload my own target and FR curves, for example, if I don’t like the ones already provided. And can use any curve I want for purposes of compensation.
If, for example, I want to see how the response of my headphone compares to another headphone, I can use either one as a compensation curve to show just the differences, which effectively circumvents the sine illusion you mentioned above.
I can also see the difference in response between my headphone and a group of other headphones by using my headphone as the compensation curve. These can be powerful tools for visualizing, analyzing, and comparing the differences in responses imo.
I still do alot of compensation right inside EAPO’s Configuration Editor though, using its stacking feature.
1 Like
Yes, and it has to be said that this is a very useful thing to be able to do, particularly if you know what you’re doing. But we are currently moving towards headphone data visualizations that aim to provide a more complete picture of the data than the whole lines vs targets stuff squiglink is well set up to do.
3 Likes
Hi Resolve, thanks for the recent tier list you posted. From what I understand, the highest headphones are only ranked around 7, while a 10/10 can only be achieved by EQ. That is quite a large gap, and I’m curious how you achieved such a 10/10 sound. For instance, which headphone did you EQ to get the result, and could you share the EQ? I understand it likely won’t work for anyone else because of HRTF and such, but I would still be interested in seeing what you adjusted.
I’m also curious on how you would describe such a 10/10 sound, subjectively. My current understanding is that certain deviations from Harman (what I consider neutral) must be made, in order for subjective effects to be achieved. For example, a 2k dip correlating with a larger soundstage. So I imagine that an hd600 EQ’d to have a large soundstage, would not be able to retain the perfect mids it has. In the same way, every EQ I’ve done to my susvara seems to harm the sound, such a bass shelf causing a loss in clarity. TLDR is that I find frequency response to be a balancing act, and you can’t have everything at once. Is that the case with a perfect EQ? With a 10/10 headphone, are you able to get perfect mids, a large soundstage, and a super detailed presentation all at once?
It’s worth noting that individual scores ended up higher than 7/10 without EQ. The reality is just that no headphone achieves higher when you consider all aspects of sound quality. As to how to achieve 10/10, yes I think that does necessarily require personalization. But we’re not yet at the threshold where that’s the core part of the conversation. Headphones can be quite a bit better still before even getting to the “it needs personlalization” stage.
As to which headphones I EQ to get 10/10, it’s honestly doable with most headphones. But I prioritize designs that are A) easier to achieve it with, and B) good for other aspects, like form factor and comfort. So I’d be happy to do it with a Sony MV1, but also happy with an Arya Unveiled - I can see an argument for either one, for different reasons. But both can achieve 10/10.
So we really shouldn’t think of Harman as ‘neutral’ in any strict sense. It’s one example of Neutral, but neutral as a concept really is more about a range of potential sound quality and proportional bass and treble. Harman gives you one version of that.
But as to your conjecture about deviations from that achieving certain tasteful colorations that impart various subjective effects, yes that is definitely the case. The question is though… colorations from what? The reality we don’t often talk about is that the defining characteristics of the Harman target are in fact colorations. These just happen to be wideband colorations we can be confident people like.
So I’d say on a more fine-grained analysis, you’d be looking at A) colorations from your brain’s expectation of ‘normal’ (which is based on your anatomy), and B) frequency response relationships that impart various effects that are going to work well for the music you’re putting through it. So for example, the combination of a slightly forward 1.5khz and low mid / upper bass bloom you get on the Focal headphones creates this intense hard hitting sensation, and works incredibly well for certain types of music because it accentuates those tones - it brings the flare, so to speak. But those features can also be heard as colorations from what the brain expects.
With respect to tradeoffs, yeah that’s absolutely right. When you’re creating these colorations to impart whatever effects, like soundstage and so on, it does require a hit to timbre. You just need to decide which aspects to sound quality matter more to you and where those tradeoffs become a problem. But again, that’s the same for all colorations really. You want more bass… but at a certain point that’s going to negatively impact detail and clarity. It all needs to be considered together, and as you say it’s a balancing act.
Really it is up to the individual, it’s one of the reasons why the community has to be willing to accept the vibes nature of this whole space, and reject any notions of audiophile authority. It’s worth considering headphones with a response that reasonably follows an ear transfer function of some kind, DF condition ideally, and then has proportional bass and treble to a degree that we know people enjoy, and then go from there. But there’s no guarantee it’ll sound better to you than something that has more meaningful colorations.
2 Likes
Thanks for the response. I must say I am very surprised you say headphones can get better, even without personalization. From my understanding, we figured out what “generally good” (harman) was a long time ago, and the rest is different forms of coloration. Which of course, are dependent on the individual. What exactly do you mean when you say they can get quite a bit better?
Related to that, I’m also still curious how how you would describe a 10/10 headphone. What exactly is the difference between a 7/10 and 10/10 headphone? I understand that the headphones lower on your list usually have some treble peak, rolled off bass, or other problem in the frequency response. But take something like the D1, which you describe “A neutral headphone that’s the closest thing to an upgraded HD 600, benefits from a bass boost” in your list. I can’t imagine that a little lack of bass makes the headphone drop to a 6.5/10 on your list. Point is, you seem to describe a massive gap between EQ’d headphones and nonEQ’d headphones. As someone who has only done EQ profiles and some EQ by ear, I certainly don’t have that same experience. To me, EQ is simply a tool to fine-tune your headphone, and yeah it can fix horrible sounding stuff, but making something sound better than a susvara? Yeah no, I don’t have the ability to do that, hell I don’t even understand what that would sound like.
2 Likes
There are a number of issues with Harman, one is that people have made use of it out of context, but another is that they assumed headphone behavior would trend more similarly between rig ears and human ears than it actually does.
The research outcomes are still valuable, but we shouldn’t be using the target the way people have been, in large part because the rig ears people use to measure relative to the Harman target aren’t the same as what was used in the research, and therefore cause headphone behavior to trend in a unique way that few actually understand. So in short, nobody actually knows what the Harman target is supposed to sound like. They think they do, but they aren’t making comparisons based on measurements done on the right ears.
This is made worse due to other factors like high acoustic impedance, which creates a big problem when trying to evaluate closed back headphones relative to the Harman target like the way we see commonly done. The graph is not a very accurate depiction of headphone sound, and as a result provide a poor guide for EQ.
It’s typically an unveiling of sorts. But I also don’t think there’s a linear progression from 7/10 to 10/10, people can like things for different reasons, and we know that the same people can rank very different sound signatures similarly. So I don’t think every headphone EQ’d to 10/10 for me would sound the same.
I think fixing the bass on the HD 600 would take it to like an 8.5 overall, something like that. I forget how I ranked the other aspects, though I imagine rather highly. But there’s still room for some improvement in other regions, albeit minor in comparison.
Yeah it’s not easy, particularly past a certain point. But this is also why there is genuinely a benefit in buying headphones that are already exceptional in the treble. It’s way easier to bring those to 10/10 than anything where you have to use all kinds of fine-grained adjustments in the treble.
1 Like
Maybe you’ve already done this, but that would be a great topic or episode for a live stream – how to do basic PEQ of the HD600, or something like this: thy this filter, how does that sound? etc.
“The EQ Seminar”
2 Likes
I would be interested in seeing what a personalized EQ preset that took a headphone to 10/10 looks like. So that this preset can’t be “abused”, you don’t have to tell us the headphone model. Although I would like to know that as well. I’m just curious where and to what extent the changes are.
You mean like the FR result for a given headphone I’ve EQ’d? Or the EQ filters themselves. Because the latter will heavily depend on the headphone.
The filters. That’s why I’d like to know the headphone. But the result would also be interesting.
EDIT: For something relatively easy for you to EQ like the Arya Unveiled
I tend to post those filters on the forum threads when I make them, though yes in large part that’s for filters that adjust ranges I can be confident will track similarly for different people. So in that case it’d be something like reducing 5khz and maybe boosting the bass by a bit - I don’t happen to have that headphone in anymore (sadly). With that said, that’s probably still just 8 or 9 out of 10 for me personally. I’d have to get into more specifics for myself past that point. So to that end, my perspective is that we’re not yet at 8 or 9 out of 10 for headphones without personalization, meaning there’s definitely room for improvement for this industry as a whole, but it would have to be some crazy stroke of luck for a 10/10 headphone to exist for me without EQ. Even more so with IEMs.
I get it. I was just curious, since you are much more experienced at this than I am, what the filters for a headphone you have EQ’d to yourself look like. Like how many and what the gain and Q factor are. Are you going in and just shifting a narrow freq range by 0.5db or are you using LS and HS filters at a particular frequency to balance the timbre to your preference? Etc.
I guess what I’m looking for is a feel for just how precise I would need to be to get my own headphones to a 10. Because I feel like I have already done that with my Empyrean II, but I’m still using pretty broad filters. How would I know I’ve maximized the Empy 2 or any other headphone I own?
Hey if broad filters is all you need, then that’s ideal.
But yeah I do use a range of Q values. Mostly just peak filters. Here’s an example for the 99 Classics V2 (not to say this 10/10 for me yet):
Of course! Yeah. But how do I know that is all I need? How do I know what the limit of an Empyrean II is in terms of EQ’d sound quality? I don’t know what I don’t know, y’know?
Full disclosure I recently purchased the Warwick Acoustics Bravura (my humble review in progress) and it has sort of broken my brain on what I thought was possible, particularly around overall sound clarity. It is such a significant upgrade to the Empyrean II (and every other headphone I have A/B’d it with) in this regard - without even using EQ - that I’m scratching my head as to how it does it. I’m not a big “driver story” believer. In fact I was fully on board with in situ FR is all there is. But this new experience does give me pause. So I start to think that there must be something I’m missing with my Empyrean II EQ.
Needless to say a video guide on how to implement personalized EQ is greatly anticipated. 
1 Like
Ok I just had a thought about this. Maybe in situ FR is all there is BUT there is a limit to what EQ can correct. For example, let’s assume that the Duo pads on the Empyrean II tend to cause some clarity loss vs the Bravura’s solid leather pads due to how the higher frequencies interact with the more absorbent and less smooth Alcantara. It is conceivable to me that this would be difficult if not impossible to correct using EQ.
I tend think with stuff like that a more likely explanation is to do with in-situ response varying more greatly than expected. Like I think we’re all flying more blind than we realize with FR in situ, and if you think about it, there are an endless amount of FR combinations going on, so it’s hard to say “fix this bit and it’ll be all good”. But I’d say keep trying and see what you come up with.
Another possibility to consider is that certain colorations can be more positively received, and really that may be what you’re looking for. And that’s where I’m at my own limits as well. I have a strong sense of a few key ones for myself but… really hard to know which kind of flare is going to yield a firework show vs a shotgun blast for people.
I have at least 3 colorations that I am aware of and enjoy. I often manually apply them using EQ to headphones that do not have them. In general it doesn’t work as well as I expect but I always learn something.
I have no idea what the “clarity” coloration is, but I intend to investigate using what means I have.
1 Like
@AudioTool : Did you try the Elite hybrid pads on the Empyrean II? In my opinion this combo sounds better than the duo pads.
@Resolve : Luxsin does the next step and offers AI-based EQ implementation. I am curious what you think about it.
2 Likes