TRUTHEAR x Crinacle Zero IEM Review

[DanielT]

They need a lot of power to get them going out from my mojo2, I’m on - 22db (green on mojo2) and could go higher without much trouble, my timeless usually hovers around -42db and vulkan around -36db so they’re quite power hungry. Staging/width is on par to the few iem’s that I’ve listened to so far and I haven’t experimented with EQ yet.

Hm, then a reasonable USB C dongle may be needed, which has decent power, gain. I intend to use my TRUTHEAR x Crinacle Zero together with my mobile phone.At the moment I don't have any good, or ok,sensible USB C dongle.

Damn, now I'm going to have to buy more. I didn't think about that. After all, I had introduced a purchase freeze for more HiFi this year. Never mind, there may be another exception. Some USB C dongle with reasonably good performance, gain/power that goes well together with TRUTHEAR x Crinacle Zero , which costs around $50. Tips on that are welcome.

Another tip I took note of. Do not buy any USB C dongle within the EU, because such a dongle CAN unfortunately then have an amplification, gain brake built-in. Some strange EU rule I read about. So importing from China or some other country outside the EU.
(I know it sounds unbelievable, but it wouldn't surprise me if the bureaucrats within the EU came up with something like that)

Edit:
Maybe this one?

"230mW@32Ω, 54mW@300Ω."

I think I found the perfect USB C headphone dongle...

The Moondrop Dawn - https://shenzhenaudio.com/products/moondrop-dawn-3-5mm-dual-cs43131-chip-portable-usb-dac-amp-pcm-768khz-dsd256-type-c-headphone-amplifier-1 These specifications sound amazing! ...better than my Activo CT10 DAC. The CT10 DAP also has 2v output. I know the specs are...

audiosciencereview.com

 

[DanTheMan]

Fundamentally, it started off based around neutrality. It's based on an Anechoic Flat speaker being measured at the eardrum of a dummy head in the Harman Listening Room to emulate an equilateral triangle stereo listening position. They then allowed study participants access to tone controls in bass & treble to manipulate the target curve (which was created from the process in the previous sentence), to adjust the sound to their preference whilst listening to a variety of reference music tracks. The Harman Headphone Curve is an averaged result of that process. That's my understanding. So, the Harman Targets do have neutrality as their fundamental, but then there is some preference added on top of that - and for good reason because you lack the tactile bass response that you get from speakers and to a lesser degree from over ear headphones - which is why you have more bass than "neutral" in headphones and even more bass in IEM's vs headphones (as it compensates for the lack of "tactile" bass). So it's based on neutrality and then converted for preference......but there's a good argument to say that this preference is also a reflection of true neutrality/experience as of course you increasingly lack the tactile element of bass as you progress from speakers through to headphones & then through to IEM's. The way I see it, the Harman Curves for both speakers/headphones/(and yet for me to try IEM's) are actually our best possible current reference for neutrality as well as preference.

But it doesn’t appear to be based on neutrality of a stereo set of loudspeakers and stereo measures dramatically different than mono.

 

[DanTheMan]

Unfortunately it wasn't exactly like that, but the basis was the response at the eardrum of a good loudspeaker in an average room that was equalised to measure flat at the LP (imho mistake #1) and then the two predefined (frequency and Q - mistake/limitation #2) shelving filters for bass and treble were adjusted by listeners to taste listening few test tracks like "Cousin Dupree" and "Bird on a Fire" (limitation #3). Luckily the positions of the shelving filters were quite well chosen so that the final response doesn't differ too much from the response of a good loudspeaker without EQ, but there are some differences.

This is also part of the reason why I don’t understand why we are making such a big deal out of this curve. Thank you for explaining this. If any of us have listened to a lot of stereo equipment over the years, it’s obvious that the Harman Preference curve doesn’t sound right. When you read about how it was done and think about how they could have done it better, it’s apparent why that is. Yes, some of it is the lack of tactility which all headphones suffer from, but some of it is just the foundation that couldn’t be remedied by to 2 EQ sliders. Still, it’s a start. Something that’s been needed for a long time.

 

[DavidEdwinAston]

Please can I comment that my new Moon drop Chu, sounded extremely quiet when powered by the Artera Platinum monoblok combination. The same volume produced perhaps 55/60 dB listening through the loadspeakers.
Does this sound normal?

 

[MayaTlab]

To illustrate further, the two IEM targets Harman tested, despite only differing by a couple of dB, on average over all tests (not just cherry-picking one test) were given a scaled preference rating by listeners 10 points apart, as can be seen below by the two highest rated data points at 100 (the Harman target) and 90 ('Harman target 2'):

For practical reasons, for the 2017 IEM articles (and later over-ear articles), instead of asking listeners to score all 32 curves (including the IE target) during the same session, they separated them in five sessions containing each 6 curves (including in one session “IE target 2”) and the same two anchors in all sessions, a high one (IE target) and a low anchor. The use of the anchors is to stabilise the use of the scale across all five sessions, as Harman already knew that when presented with different sets of alternatives listeners would use the scale differently. The thing though : the use of anchors doesn’t fully eliminate this variation. The high anchor (the IE target) and low anchors vary in absolute terms by around more or less five points on a 100 points scale across the five sessions (before the scaling shown in your aggregated plot, which is not the same as the one published in the article BTW). The same can be seen in the over-ears article.

Hence why when it comes to assessing how the IE target was preferred to target 2 it makes sense to only "pick" the fifth session, where listeners were able to directly compare them and instantly switch between them at will, to avoid that phenomenon. And rated them similarly, regardless of training.

AES E-Library » A Statistical Model that Predicts Listeners' Preference Ratings of In-Ear Headphones: Part 1—Listening Test Results and Acoustic Measurements

A series of controlled listening tests were conducted on 30 different models of in-ear (IE) headphones to measure their relative sound quality. A total of 71 listeners both trained and untrained rated the headphones on a 100-point preference scale using a multiple stimulus method with a hidden...

www.aes.org

So obviously even small differences in frequency response can produce a non-insignificant change in preference.

Nah. What you’re seeing here most likely has little to do with how differently they sounded and more likely everything to do with the arbitrary methodological choices Harman made when aggregating and scaling the results from the five sessions - arbitrary choices that were unavoidable given the method used anyway. It is not precised in the article whether they averaged the scores from the IE target and low anchors across all five sessions or randomly picked one from a particular sessions, but either could have resulted in the IE target scoring higher in the aggregated plot even though it didn’t in the direct comparison during the same session.

The reason for being so meticulous here when looking at frequency responses and variables actually under our control like EQs is precisely because of all the potential inaccuracies in variables we cannot control - you do not want to compound errors. The fact that some repeatedly fail to grasp this, on a science forum, is pretty worrying.

The methods Harman used throughout their research in target curves, as well as the act of measuring a pair of headphones on a 711 coupler and comparing it to the target, already are compendiums of unavoidable errors in the first place, occasionally arising from arbitrary methodological choices, not just measurement precision / accuracy - and this is not a dig at the research in the slightest as I don’t see how it could be any different. Cf the paragraphs above for one (albeit I guess utterly negligible in light of others) source of potential error.

An analogy: you're tasked with making a table, but you're given a tape measure that you've been told is only accurate to ± 2mm. Do you say, 'ah that's not very accurate, so I might as well just do it all by eye', or do you make sure you measure every piece of wood to cut as carefully and accurately as you can so you don't introduce any more variance?

A more appropriate analogy is to consider that 5 people are asked to measure 5 tables with 5 different tapes and go through all the permutations. You get a +/- X mm constant offset for each person, each table and each tape, and on top of that you get a random +/- Ymm offset for each measure.

How much exactly these offsets vary is not quite yet fully known but some partial solid data is already available, just like we already have a pretty decent idea on leakage issues at lower frequencies for over-ears and how, for example it varies across different types of over-ears (from a minor issue to a major one).

 

[Jimbob54]

Please can I comment that my new Moon drop Chu, sounded extremely quiet when powered by the Artera Platinum monoblok combination. The same volume produced perhaps 55/60 dB listening through the loadspeakers.
Does this sound normal?

How are the chu connected to the amp?

 

[Robbo99999]

Unfortunately it wasn't exactly like that, but the basis was the response at the eardrum of a good loudspeaker in an average room that was equalised to measure flat at the LP (imho mistake #1) and then the two predefined (frequency and Q - mistake/limitation #2) shelving filters for bass and treble were adjusted by listeners to taste listening few test tracks like "Cousin Dupree" and "Bird on a Fire" (limitation #3). Luckily the positions of the shelving filters were quite well chosen so that the final response doesn't differ too much from the response of a good loudspeaker without EQ, but there are some differences.

Yes, you're right about the "equalised to measure flat the LP" part, which is actually the following graph being the measurement at the eardrum of the dummy head in their study:

And it was an F208 speaker in their Listening Room which they used as the reference speaker that they EQ'd to a flat in-room response, and following is their F208 in their listening room (albeit it's not EQ'd flat in the following graph):

But interestingly if you apply the exact slope of the black line in Figure 19 to the Curve shown in Figure 5 you get an exact representation of the 2013 Harman Curve - I did an exercise in REW one time where I traced the slope of that black line and applied it to the Curve in Figure 5 and it completely overlays onto the 2013 Harman Curve. Of course there's more bass & treble in the 2018 Harman Curve, and following a pic of the different Harman Curves for reference:

I'm not sure I agree with you on the fronts where you say that there were mistakes re their decisions, but that's up for debate.

 

[oscar_dziki]

I was about to buy these IEMs but declined on the ground that I don't want to be misidentified as a pedophile.

Okay guys, I'm sorry but this whole pedo aspect sounds weird to me. How can you even tell that this manga girl is a minor? It seems to me a little overblown.

 

[Robbo99999]

No, not an anechoic room, this is false, a room as reverberant as a typical studio.

I didn't say that. But my see previous post directly above this one, and my conversation with thewas, we've cleared that up.

 

[DanTheMan]

Yes, you're right about the "equalised to measure flat the LP" part, which is actually the following graph being the measurement at the eardrum of the dummy head in their study:

And it was an F208 speaker in their Listening Room which they used as the reference speaker that they EQ'd to a flat in-room response, and following is their F208 in their listening room (albeit it's not EQ'd flat in the following graph):

But interestingly if you apply the exact slope of the black line in Figure 19 to the Curve shown in Figure 5 you get an exact representation of the 2013 Harman Curve - I did an exercise in REW one time where I traced the slope of that black line and applied it to the Curve in Figure 5 and it completely overlays onto the 2013 Harman Curve. Of course there's more bass & treble in the 2018 Harman Curve, and following a pic of the different Harman Curves for reference:
View attachment 231690

I'm not sure I agree with you on the fronts where you say that there were mistakes re their decisions, but that's up for debate.

I know you weren’t responding to me, but the HD 800 has the “face tweeter” reputation.

It would be harder to argue that there weren’t mistakes. Particularly with all the redos. You’d have to skip factoring in the 1dB/oct slope above the bass bump to align those curves.

 

[Robbo99999]

how did you do that EQ? There’s absolutely no way that’s the balance you should be hearing with your studio monitors in standard locations:

View attachment 231673View attachment 231674

View attachment 231657
That is the response you get when 2 speakers EQed to fairly flat at the listening position. The Cancellation at 20kHz is proof that the microphones are centrally placed. That response changes to this response that I superimposed over the last graph.

View attachment 231658
The shape of the blocked ear pinna gain correlate very well to Dr. Toole’s and Sank’s data of pinna gain at 22.5 degrees even though my monitors are at 30 degrees. Even head-on, results are not that different
View attachment 231659
They are not alone:
View attachment 231660View attachment 231661View attachment 231672
etc..
Then when you add in the head shadow:

View attachment 231665
And add a head shadow from the other direction to make a stereo pair:
View attachment 231664


you take a frequency response that looks like this:View attachment 231657
And it becomes this:
View attachment 231658
Harman’s curve looks to be based on a mono speaker:
View attachment 231659
But without the air and a lot more low end.
View attachment 231668
Maybe their missing 6dB, or whatever it was would have been found if they had used a stereo source to begin with?

So why is it that we are so hung up on a graph that more than 2/3 of a lot of people preferred over a different graph if neither graphs are closely related to a stereo triangle? I’m absolutely missing the logic behind it. I made a quick sketch of the regions of the graph to clarify it a little bit, but basically approximately 1,500Hz, we are going to hear both loudspeakers equally. You can tell by my graph that it’s not black and white, but a smooth transition from one condition to the next. Above there, the ipsilateral loudspeaker will be the more dominant sound source. This is a science forum. We should be able to discuss this here with some fluency.

View attachment 231678

Or this scary one! (Thanks @Robbo99999 )

View attachment 231681

I know you have concerns that you think the Harman Headphone Curve doesn't seem to match your results of your own blocked ear canal mic measurement results of your own speakers in your own room, and along with the various extrapolations you've done to try to equate the two. To be honest I'm not expert enough (& with the time & dedication I have today) to attempt to explain the differences you see vs the Harman Headphone Curve - not beyond the generic answers of:
(1) It's your room & your speakers, not the speakers and room used in the Harman Study
(2) It's your head and your outer ears (your partial HRTF) vs that of the dummy head mannequins head and outer ears and ear canal (it's full HRTF)
(3) I don't know how accurately you can translate your closed canal measurements to what your own eardrum measurements would be.

In terms of someone who would be able to quickly digest what you're trying to show and who would probably find it easy to give a reply to your questions I can think of a guy called @Mad_Economist who is quite an expert on this, I think he has dabbled in designing his own headphones as a hobby and has consequently read & accumulated a whole host of knowledge and information that he seems to have at his fingertips to answer such questions as yours. I don't see him much around on ASR anymore, but I still think he occasionally drops in. He has also studied other headphone work/research beyond Harman. If Mad Economist can't answer you, then there might be a few other members that can easily answer your questions.

But a quick answer to one specific question you asked me, you said:
"how did you do that EQ? There’s absolutely no way that’s the balance you should be hearing with your studio monitors in standard locations"
You were referring to the following graph I showed:

All that is me doing a slight tweak to the Harman Headphone Curve to tweak it to sound more like my studio monitors - so this was subjectively listening to my reference tracks and flipping quickly between headphone and studio monitors and adjusting the EQ of the headphones (with a Harman Headphone EQ as the starting point) to match closer to what I was hearing through the studio monitors (I did have them in an equilateral triangle). As you can see the only thing I changed was giving it a bit more pinna gain and for that pinna gain to start happening just a little earlier. The blue line is the Harman Headphone Curve, and the highlighted line is my equalised headphone. So that's not an unreasonable tweak & also not a massive tweak, I'm not sure how you can say "there's absolutely no way that's the balance you should be hearing with your studio monitors".

 

[oscar_dziki]

So far seems like a very solid buy at £50, I doubt anyone could find anything to complain about with sound quality from a dac/amp but I’m not sure how well a phone jack would do to power them.

I just tried them on my Samsung a52s and then switched quickly to Meizu 6 pro - also a phone but with a more serious DAC that can handle even 250-ohm cans quite nicely. The difference was shocking, On Samsung, they sound so muddy... like completely different earphones. If I only heard them on Samsung I would be returning them.

 

[DavidEdwinAston]

How are the chu connected to the amp?

A qed speaker/headphone switching box, four metre cable then 6.35 adaptor plug. The qed presumably out from the Artera analogue preamp, rather than the power amplifiers! Thanks Jimbo

 

[DanTheMan]

I know you have concerns that you think the Harman Headphone Curve doesn't seem to match your results of your own blocked ear canal mic measurement results of your own speakers in your own room, and along with the various extrapolations you've done to try to equate the two. To be honest I'm not expert enough (& with the time & dedication I have today) to attempt to explain the differences you see vs the Harman Headphone Curve - not beyond the generic answers of:
(1) It's your room & your speakers, not the speakers and room used in the Harman Study
(2) It's your head and your outer ears (your partial HRTF) vs that of the dummy head mannequins head and outer ears and ear canal (it's full HRTF)
(3) I don't know how accurately you can translate your closed canal measurements to what your own eardrum measurements would be.

In terms of someone who would be able to quickly digest what you're trying to show and who would probably find it easy to give a reply to your questions I can think of a guy called @Mad_Economist who is quite an expert on this, I think he has dabbled in designing his own headphones as a hobby and has consequently read & accumulated a whole host of knowledge and information that he seems to have at his fingertips to answer such questions as yours. I don't see him much around on ASR anymore, but I still think he occasionally drops in. He has also studied other headphone work/research beyond Harman. If Mad Economist can't answer you, then there might be a few other members that can easily answer your questions.

But a quick answer to one specific question you asked me, you said:
"how did you do that EQ? There’s absolutely no way that’s the balance you should be hearing with your studio monitors in standard locations"
You were referring to the following graph I showed:
View attachment 231691
All that is me doing a slight tweak to the Harman Headphone Curve to tweak it to sound more like my studio monitors - so this was subjectively listening to my reference tracks and flipping quickly between headphone and studio monitors and adjusting the EQ of the headphones to match closer to what I was hearing through the studio monitors (I did have them in an equilateral triangle). As you can see the only thing I changed was giving it a bit more pinna gain and for that pinna gain to start happening just a little earlier. The blue line is the Harman Headphone Curve, and the highlighted line is my equalised headphone. So that's not an unreasonable tweak & also not a massive tweak, I'm not sure how you can say "there's absolutely no way that's the balance you should be hearing with your studio monitors".

The only real question I have is “why should we care about this curve?” As a lot of science minded folk, why would we uphold this curve as the final answer? My data is good and it matches the the luminaries for that matter. That’s why I was showing it—to demonstrate that my data aligns and I’m not just some lunatic barking at the moon. I’m certainly not questioning it. I’m supporting it with other’s data. I build headphones as well. Listening to a set of them right now. I’m not new to this In case that isn’t obvious. I took that data nearly a decade ago.

FWIW, I’m sure that this curve isn’t horrible, I’m just as sure it could be improved and I’m demonstrating a fundamental issue with it. I have a few headphones that natively measure close and at the push of a button, I have many that match it. I know if a headphone measures like this, I don’t want it. Firstly because it won’t bring me anything new and secondly because I’m not a big fan of what it will bring me. To be honest, I no longer have use for IEMs either, but that’s outside the realm of why I wouldn’t want this. I’d much rather own a Stealth, HD650, etc… Something that doesn’t get this where bass is more audible.

Even Amir said he preferred this 70% of the time (or something like that).

 

[Robbo99999]

The only real question I have is “why should we care about this curve?” As a lot of science minded folk, why would we uphold this curve as the final answer? My data is good and it matches the the luminaries for that matter. That’s why I was showing it—to demonstrate that my data aligns and I’m not just some lunatic barking at the moon. I’m certainly not questioning it. I’m supporting it with other’s data. I build headphones as well. Listening to a set of them right now. I’m not new to this In case that isn’t obvious. I took that data nearly a decade ago.

FWIW, I’m sure that this curve isn’t horrible, I’m just as sure it could be improved and I’m demonstrating a fundamental issue with it. I have a few headphones that natively measure close and at the push of a button, I have many that match it. I know if a headphone measures like this, I don’t want it. Firstly because it won’t bring me anything new and secondly because I’m not a big fan of what it will bring me. To be honest, I no longer have use for IEMs either, but that’s outside the realm of why I wouldn’t want this. I’d much rather own a Stealth, HD650, etc… Something that doesn’t get this where bass is more audible.

Well, it sounds like you've put a lot of work into it, and you certainly seem to know what kind of target curve you like, which is fine......so in fact you're in a good position to be able to tune headphones to the sound that you require - that doesn't have to be Harman Headphone Curve. Not everyone can be such a dedicated hobbyist such as yourself though, so they have to rely on "best accepted standards", and the Harman Headphone Curve is currently such a thing. For what it's worth, I think it's a very solid curve, and works very well for me - I mean it's not that surprising given the curve was created by starting off with objectivity as starting point and then adding some preference on top of that based on a number of studies that Harman did using a fairly large number of participants (don't ask me how many, lol). I personally think that a more accessible version of headphone personalised DSP in the same vein as Smyth Realizer and The Impulcifier Project are the future of headphone listening, thereby accurately incorporating your personalised HRTF and emulating real speakers in a room to good success - but for now we have the Headphone Harman Curve until stuff like that becomes affordable and/or easily accessible without "hassles".

 

[buz]

Seeing that this one basically needs no eq, I am tempted to just stick a usb c cable dac on it permanently kl(otherwise would use my qudelix). Which one would be fitting? Apple?

 

[RHO]

Okay guys, I'm sorry but this whole pedo aspect sounds weird to me. How can you even tell that this manga girl is a minor? It seems to me a little overblown.

Yeah, I don't get it either.
It's just a print of an uninteresting drawing on a piece of thin cardboard. What the f is all the fuss about?

 

[DanTheMan]

Well, it sounds like you've put a lot of work into it, and you certainly seem to know what kind of target curve you like, which is fine......so in fact you're in a good position to be able to tune headphones to the sound that you require - that doesn't have to be Harman Headphone Curve. Not everyone can be such a dedicated hobbyist such as yourself though, so they have to rely on "best accepted standards", and the Harman Headphone Curve is currently such a thing. For what it's worth, I think it's a very solid curve, and works very well for me - I mean it's not that surprising given the curve was created by starting off with objectivity as starting point and then adding some preference on top of that based on a number of studies that Harman did using a fairly large number of participants (don't ask me how many, lol). I personally think that a more accessible version of headphone personalised DSP in the same vein as Smyth Realizer and The Impulcifier Project are the future of headphone listening, thereby accurately incorporating your personalised HRTF and emulating real speakers in a room to good success - but for now we have the Headphone Harman Curve until stuff like that becomes affordable and/or easily accessible without "hassles".

You are exactly the type of case why I wish they had made a better effort to start with. Now we are stuck with this as a design goal. It’s hard for the community to take you serious unless you conform and that means consumers are missing out. They could have made a neutral curve and then a preference curve based off that. Instead they went back to the early days of audio—mono—and started from there. There are some legit advantages to the approach. I mean when 2 humans speak to each other, they are mono a mono. Same goes for a girl with a guitar…. It’s just that most, if not all, recordings are mixed and mastered using stereo and so manipulated for the format. Still, mono instruments may actually do better Harmanized in a sense and people can actually prefer it. Life and headphone curves are a balancing act. Just conform and everyone will like you.

 

[Biblob]

Harman named after the founder Sidney Harman.
I shrugged btw. IE2019v2 sounds shrill. IEF plus bass sounds better.
OE2018 is nice though.

Any IEM recommendations which follow this curve naturally?

 

[staticV3]

Any IEM recommendations which follow this curve naturally?

Monarch MKII, Thieaudio Clairvoyance, Moondrop S8
or maybe wait for this secret project to be revealed/released: