OP
- Thread Starter
- #121
You got it. Will never partner again. No two people see eye to eye on all of these things.
-
WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
ASR Headphone Testing and BK 5128 Hats Measurement System
- Thread starter amirm
- Start date
Just to clarify, is your endorsement a yes, you will test one of these if I send it? Seems like the Bose would be the best choice based on popularity.
If the Harman target is as bad as you claim, how did Dr. Sean Olive's headphone preference formula (which is based on adherence to this target) achieve the very high 0.91 correlation between predicted and actual preference of in-ear headphones (and the same high 0.86 correlation for over-ear headphones as his speaker formula)? And how can you possibly say the Harman target is "bad and inconsistent" yet Moondrop IEMs are "good shit", when the latter follow the Harman in-ear target more closely than the vast majority of IEMs available?
Oratory is one of the nicest and most knowledgeable guys in this hobby in my experience by the way. He's also a professional acoustic engineer, with degrees in technical physics and energy & measurement technology, who's specific job is to measure headphones as part of the research his company does in Austria (which also includes some ex-AKG engineers as employees). He measures people's headphones that they send to him in his spare time and asks for absolutely nothing in return, as well as calculating the Olive preference rating for them and producing high quality parametric EQ profiles for each headphone he measures, and even making custom profiles on request (e.g. graphic EQ, or the number of bands of your choice). He was an absolute pleasure to deal with when sending my headphones to him and asking for custom EQ profiles, which sound fantastic. He also takes the time to explain in a patient, clear and detailed manner the often complex workings of headphone electroacoustics and measurement to anyone who asks on Reddit. And he does all this without the hubris and condescension that seems to be rife in many corners of the audio world for some reason. Much like Tyll of Innerfidelity before him, I would say he's done more for the headphone community than pretty much anyone else in recent years, not least in dispelling pseudoscience myths and popularizing the science and objective measurement of headphones, and providing an incredibly useful and growing database of hundreds of headphone measurements and EQ profiles. If you claim a professional's measurements are "one of the worst in terms of truthfulness", you need to provide specific hard evidence of this, and exactly what would constitute truthfulness in this context. Otherwise it just seems like baseless insinuation due to some personal beef you've had with him.
I have a question. Let's assume our target is harman curve. If the measurement rig oratory uses is the same as the one used to produce harman curve, does it make oratory's measurement better than other measurement rig?
You have to grasp the concept of what's the scale of things we are talking about. Yes Harman target is not that bad.
The thing is Moondrop iems are CLOSER than majority of other iems to harman target doesn't mean harman target is good. IEMs that are tuned to Harman target can sound drastically different. It also means majority of IEMs on the market sound worse than dog shit. (you can feel the difference between pretty bad and dog shit right?) There are just too many things. Also about consistency. For speakers that are within +-3db 90% consistency is good. For headphones and iems that are worse than +-5db I need 99% or 100% prediction accuracy. Headphones and earphones sound so different it should take only 2 second to accurately evaluate.
If you think about it, harman target is tuned every couple of years. The accumulated error is so large it's not even funny. The newest target tuned more than 4db in some areas. And measurements have no truthful indication of how shallow inserting iem sound around and over 6khz. Making the measurements pretty much useless for scoring. A peak caused by ear canal and the iem will ruin everything. Measurement equipment is only accurate when deep inserting like er4. That's the volume it's designed for. Shallow inserting type will not be accurately measured.
Then you can take speakers as example, all speakers are tuned to flat right? Many are +-3db within that range but do they sound the same? Not remotely. Similar probably. Then go back and look at how many iems or headphones can hit target +-3db.. Let alone +-1.5db....
That's why I love etymotic, they have done pretty much everything almost 30 years ago. Er4 is still the best sounding iem. It's freaking released in 1991.
PS: I'm not even mentioning the IE target of harman target which is simply shit. OE is ok.
For the last part. Whenever I see he was talking, he's always on a high horse.
For accuracy and truthfulness? I have said. hd58x and er4 just look that them. Do they look normal?
There are other methods of measuring frequency response. Best way is to use actual human and possibly human hearing itself. If there's roll off in the highs we can hear it comparing to flat speakers just use sine sweep. It's best to identify peaks and valleys. Put on your headphones and compare the peaks and valleys to Oratory's measurements. I don't think anyone that can actually hear would use his eq. Many measurements are simply wrong..... I know of course it's getting better but I do not check his measurements for new headphones or iems. That tells something. I had no real elevated beef with him. But after a few times reading his reply to me I just don't like the guy. He can't take in advice or even discuss about situation. Na.
Harman target's goal should not be to making best iem/headphones. What it is for is to set a target to manufacturers that tune their stuff like shit. They have talked about show ****** the hd820 is. That's what it's for. Once you are close to Harman target, harman target loses its value simply. What you should really do instead of simply matching frequency response to a target in artificial ears. You need to listen to it using music and sine sweep. You make your ears a measurement equipment. You can pick out peaks and valleys that measurement equipment fail to do. It's not like measuring amplifier. There are large differences between the artificial ears and human ears.
Matching to the target is phase 1, you need to have phase 2 and 3 to follow that up. Or it won't sound good.
Matching to the target is phase 1, you need to have phase 2 and 3 to follow that up. Or it won't sound good.
I think they use both.
There's one game that we can play to prove how inconsistent the harman target/the current measurement equipment is.
First, get a flat speaker sounding flat(to in room target) in the room without any dips or peaks. (perfect sounding speakers basically)
Second, apply EQ of harman target compensated frequency response of various headphones to the speakers.
Thrid, play the preference game and score all the different headphone EQs including the original speakers without EQ.
Then you will know the problem.
First, get a flat speaker sounding flat(to in room target) in the room without any dips or peaks. (perfect sounding speakers basically)
Second, apply EQ of harman target compensated frequency response of various headphones to the speakers.
Thrid, play the preference game and score all the different headphone EQs including the original speakers without EQ.
Then you will know the problem.
you are very contradictory to your own comments.
"Moondrop make good (expletive)."
someone tells you moondrop tunes close to harman target cureve.
"(Moondrop is) pretty bad."
Harman has only released two IE target curves, one in 2017 and one in 2019. There was a slight increase in bass shelf and upper mid-range gain. I believe it was based on adding more demographics to the testing, iirc. They've done a lot of research over the past decade that it's hard to remember which was which sometimes. It's a preference curve, so it's probably going to be a continual moving target as people age, music styles change, etc.
Etymotic has changed their sound curve over the years as well. The old ER4 and the current ER4SR/XR have slight variations as well. Mainly, there is a reduced amount of 1-3KHz and more low subbass.
I’m not knowledgeable enough to comment on this headphone testing product, but getting into headphones is a great idea. They cost less than most speakers, readers have multiple which means there won’t be a deadline to get them back to the user, lugging them up the stairs shouldn’t be a problem and shipping costs are considerably less.
‘Would you consider testing headphones only subjectively?
‘Would you consider testing headphones only subjectively?
Last edited:
To the first part, read my comment again.
To the second part, er4sr is pretty much identical to er4s. The difference is changing the placement of the nozzle on the BA driver. Other than that it's the same. The main reason of er4sr is to reduce impedance while maintaining the frequency response. The er4s used in line series resistor to tune the sound while er4sr still does it(resistor directly in the housing) it needs less resistance.
er4xr is to please the mass market. While it is a success, it sounds worse than er4sr. The driver in er4xr is ported and is based off sonion 2300 series (different than original er4 snd er4sr)(but 2300 series is essentially the ED series of knowles, a bit too much insight i know). You can't add a port and not affecting higher range, so increased damping caused darker highs in er4xr.
All in all original er4s was still the best er4sr is basically the same.
You know what? This thread kinda revived my passion about acoustics. I would probably make a new researching "stuff" that will improve the whole headphone /in ear industry.
OP
- Thread Starter
- #135
Hmmmm.
If you got a pair of good ears to trust I don't see it being a bad thing. As it bypasses all the issues in measurement equipments.Hmmmm.
OP
- Thread Starter
- #137
I think I have enough to test right now. If I purchase such a test gear, then yes, by all means. For now we are good.
- Joined
- Feb 23, 2016
- Messages
- 20,701
- Likes
- 37,442
It does and it doesn't.
He still does not have your pinna or my pinna. So results wouldn't necessarily translate.
Last edited:
OP
- Thread Starter
- #139
So we are saying science has no way of helping create a better headphone?
I would say, current measurement equipment can help creating "less bad" headphones but not enough for "better" headphones. For evaluation purposes, it depends on where you draw the line. But according to current speakers measurements and reviews I can only see less correlation between measurements and subjective results. And to be scientific we will see something like: 'your ear canals will likely be different so you may or may not have the peak' much more often.
Similar threads
- Poll
