Touched by an IEM guru, running the SBAF Dick Gauntlet, and why target curves for IEMs won't work

You just achieved SBAF immortality with " I had the honor of having a great IEM guru lecture me". Congratulations!

 

@k.e. mfw I see you're back

 

On the topic of Olive-Welti: yes it has problems, but it's already a step forward from DF tuning. These adaptations to new standards for whatever reason seem to take a long time to (1) appear, (2) become the norm and finally (3) get modified.

It suits itself well for general consumer headphones and as such is a useful tool. Blindly applying it as a correction is no good, but if you know and understand it's limitations it's still more useful than looking at nd working with DF compensated graphs.

 

not to late to grab one vansen at https://www.head-fi.org/threads/uerm-reshelled-by-perfect-seal.862265/

 

I've said it time and time again, there is no such thing as an universally valid "perfectly neutral". Please (re-)read these two posts before arguing over 1 or 2 dBs. Individual HRTFs tend to vary at least as much, and that's not even taking different listening volumes and Fletcher-Munson curves into account:

http://www.superbestaudiofriends.or...-hear-different-thread.2535/page-2#post-67954
http://www.superbestaudiofriends.or...g-headphones-by-measurements.3112/#post-91358

To give you a real-life example: @shotgunshane's and my ideas of "neutral" are very similar overall... however, his includes a bit of an Ety-bump @2-3kHz, whereas I tend to perceive such an emphasis as obtrusive. I'd never call his idea marketing influenced BS though, cause we're both experienced listeners and I'm sure our hearing / perception just works a little differently in that region.

 

Just for clarification, my argument is not against standards and targets, of course these are vital for hardware manufacturers. I'm just saying that all standards and targets for "neutral" frequency response I've seen so far are averaged curves, derived from individual measurements that vary significantly. And I haven't come across any study claiming that their target frequency response will sound "neutral" for all individual listeners.

 

What I dislike is when somebody makes false claims and fights logic with post count or date of registration.

I do think a few people around here know me and also know that I temporarily worked for a pro-audio company. Even if not, it's not like I am hiding my real name or anything. Yet I don't claim to have superior knowledge. In the contrary, I keep silent about it as to not raise false expectations and the reason I am here is to learn from others. You see, I don't want to be the guy who tells a story five times about how he visited a studio once and suddenly can redefine the word "neutral".
I also write reviews since like forever and I learned most of my vocabulary from joker back in 2009-2011. Pretty sure my terminology doesn't deviate too much from James or Shotgunshane as they were my idols back when Head-Fi was not about making cash. I do admit that it's impossible to keep track of all Head-Fi users and I admit that I was far more active on Hifi-Forum instead.
But I also introduced myself in the thread that asks for exactly that. No worries, I don't expect you to read the thread or even remember it. But it would be nice if you decided to be less of a jackass to newcomers because I can see from the thread that there are several audio students or otherwise experienced people that want to join the community. But you automatically assume you have more value than "new members" which makes you a dick.
Oh, while we're talking about dicks, unfortunately, I accidentally saw how you had a good time with hardcore gay porn in another thread a while ago. That makes you so much more sympathetic. No wonder your own people have to ban you from time to time.

I know you want to mock me with my custom tag, but actually I am proud of it. Bad_Robot was one of the best persons the hifi/IEM community ever had and he is the reason I decided to become part of the community. Not personally knowing me, he sent me five flagship IEMs of that time, only for me to experience different tunings and the capabilities of multi-BA designs. It moved me a lot and I have been sending my own gear around to users ever since. In any discussion, he never had to raise his voice or become insulting like you. Because he had reason and logic for argumentation and not your bullshit talk about how all (not Olive-Welti, but all) IEM research is crap. We know Olive-Welti is not a compensation target, it's a qualitative market research. But it is based on logic and research, especially in the midrange. Even if you do not agree with linearity, perceived neutrality or anything else, you still should use the words accordingly as that's how the terminology is used in the field of audio or by your fellow less arrogant colleagues.

According to you, all hearing aids are fundamentally wrong because you disagree with pinnae effects in IEMs. Can't you see how absurd your claims are? I guess you do, because you decided to let the anger about your own stupidity out on me. Well this is not your lucky day. I avoided you ever since I noticed you have a very different perception than me in the Andro thread months(?) ago. I gave you the last word and I apologized to the others for the shit that they had to read because you think you had to display alpha behavior. After that, I did not post in many threads I would have liked to because I could see you lurking there and picking up fights with other people already. You're nothing than an online bully to me. And I say this fully aware that we will eventually meet personally on a headphone event.

A silent observer sent me this picture: B&K measured flat in a dummy head recording and Tom's measurements of the ProPhile 8.


That's what you call shrill and horrible? You are forcing yourself and that makes your opinion not reputable. But I have an even better theory. Based on your friends, the Pepe and the guy responsible for putting SBAF on NSA's watchlist with his Japanese underage girl as an avatar, I assume society has no place for you. So you hide in the basement and you are afraid of people. That's why you don't like voices and perceive them as shrill.

Either way, I have zero respect left for your mic in a toilet roll or your opinion. I would like to appreciate the effort, but your arrogance and narcissistic ways totally blocked that possibility forever. I will continue to avoid you the best I can - how I did in the past weeks. Maybe you should try to ignore me as well. Goodbye!

 

There is a disturbance in the force: I'm sensing a ban for over-sensitivity...

 

The thing is I don't buy that we should only look at FR. The brain is way more complex than that. I'm fairly certain that two speakers with the same measured FR at the eardrum will not have the same perceived tonal balance (if they don't measure the same with an omni mic). All the target curves seem to be based on measuring FR and adjusting IEMs to a "target", but IEMs tuned by ear generally sound more correct to me. The brain knows if a sound is coming from speakers in a room or from IEMs. Otherwise the Apple BA IEMs would sound super awesome (ignoring the massive 3kHz 3rd order distortion spike and 9kHz ringing).

I guess what I'm trying to say is this: Forget all the measurements for a while. I know it's not easy to do, but try to ignore what the measurements say for a moment. Then why does stuff with a DF bump sound overly shouty and shrill to me in a direct comparison to my speaker reference (which in itself might be considered shouty by some)? I'm talking specifically about the Zero Audio Carbo Tenore and the Apple BA IEM that I have on hand. The RE-400 and UERM are much better behaved. Yes, this is an experiment that I just conducted myself. I tried to match the levels as much as I could subjectively as matching levels with a microphone doesn't work very well between IEMs and speakers (the amount of reverb influences the perceived volume level more than it changes the measurements).
I modded my RE-400 specifically to reduce the 2kHz bump. Now there's only a 2.5db bump at 2kHz on my coupler and it still sounds too forward in the upper midrange to me.
A couple subjective impressions (FR only - compared to my speakers):

• The UERM is pretty neutral overall, but over time I've realized that it's slightly too U-shaped for me with a bit too much 2kHz shout. I'd like more energy between 500 and 1.5kHz and less energy around 2kHz. Also a bit less treble energy in general.
• The RE-400 I also find quite neutral, at least the way I modded it. Quite a bit less 2kHz, smoother treble and I think just a bit more treble as well compared to stock. The dip after 2kHz annoys me, but fixing the dip is only possible with more damping material which makes the sound overly damped and lifeless. Still, the dip makes the 2kHz bump stand out too much for me. The treble level is actually closer to accurate to me, but now it's a bit too less.
• The Carbo Tenore is way too shrill, shouty and sharp. Also 10db too much bass bleeding into the lower mids. Just generally lacking the most important region of the midrange.
• The Apple IEM (newer version) is similar, but better. But the 3kHz distortion and 9kHz treble spike is a dealbreaker to me. At high levels the distortion spike can reach beyond 20%! Also the tuning is definitely still too much of a DF curve. Higher OI helps with the bass, though. The old version with less upper midrange energy is probably a bit better.

In hindsight the Brainwavz B2 that I had with the special tipless deep insertion that I always used was more neutral than my UERM (less 2kHz, less treble, more mid-midrange, better treble extension). One of the drivers crapped out on me so I'm back to using my RE 400s on the train.

 

It almost certainly does, since the HRTF is used to determine source location. Careful manipulation of the magnitude response shifts the apparent sound location, especially when it comes to sounds in the median plane. Describing HRTF would be unnecessary for loudspeakers as the brain already filters the transfer modulus out of the signal, based on other cues such as level, phase differences captured in whatever time window.

The issue with headphones and IEMs is then reconstructing the gain based on what portions of anatomy are bypassed or interacted with. The individual result has to be guessed at.