[csglinux]

Beyond common usage in audio circles, FR is magnitude AND phase as a function of frequency. As god intended...

Unfortunately, in the wider spheres beyond that, Buddha overruled both Allah and God and insisted all TLAs have to have more than 2 letters. ¯\_(ツ)_/¯

Does that not look odd to you?
I would have naively expected a mirror image of this plot...

Sorry. IR delay shift removed for Australians:

@csglinux it might be more interesting to look at either

1. A difference plot of the 2 FRs
2. Or more common to look at compensated responses vs. some target curve

In both cases, scale the Y-axis to something like +15dB to - 15dB.
Differences in FR will become much more apparent.

Great suggestion buddy - thanks! I have a feeling that's the biggest issue - things that aren't all that similar but just look that way on a particular scale (as @crinacle mentioned earlier in this thread, a lot of OEMs like to have a huge range on the y-axis to make everything look flat). Here's your suggested plot of the Tenore SPL - KSE SPL. Easy to see the audible differences here



I should probably revisit the Sonarworks software at some point. I wasn't all that impressed on first listen. It didn't make my ATH M50x sound like my HD800s. But it seems that if it's done right (tweaked to my individual models, with a sufficiently detailed EQ), it should get close. My KSE can't get as detailed as the above figure, but good suggestion there @castleofargh. Can anybody recommend PC, Mac, iOS or Android software that can EQ to this level of detail? Anybody know if you can insert your own EQ in True-Fi?

 

[buonassi]

You can recreate that spline with just about any high end EQ software. Of course you'd have to have multiple overlapping filters working in summation to achieve it. FabFilter ProQ2 would do it. Here is a little work I did for my HD600s which results in equal loudness to 1/8th octave or better (to my ears with these particular cans) when I use a controllable sine wave sweeper. The purple dot is 4khz for reference.

 

[castleofargh]

Unfortunately, in the wider spheres beyond that, Buddha overruled both Allah and God and insisted all TLAs have to have more than 2 letters. ¯\_(ツ)_/¯


Sorry. IR delay shift removed for Australians:



Great suggestion buddy - thanks! I have a feeling that's the biggest issue - things that aren't all that similar but just look that way on a particular scale (as @crinacle mentioned earlier in this thread, a lot of OEMs like to have a huge range on the y-axis to make everything look flat). Here's your suggested plot of the Tenore SPL - KSE SPL. Easy to see the audible differences here



I should probably revisit the Sonarworks software at some point. I wasn't all that impressed on first listen. It didn't make my ATH M50x sound like my HD800s. But it seems that if it's done right (tweaked to my individual models, with a sufficiently detailed EQ), it should get close. My KSE can't get as detailed as the above figure, but good suggestion there @castleofargh. Can anybody recommend PC, Mac, iOS or Android software that can EQ to this level of detail? Anybody know if you can insert your own EQ in True-Fi?

before spending money on the big boys of EQ for mixing and mastering(I have DMG Equilibrium myself, but I admit that I've watched all the tutorial vids from fab filter about a billion times. those guys rocks and Pro-Q does too), maybe just try applying what you get from the EQ section of REW into Equalizer APO on Win. I've gotten used to making my own mess with virtual cables and VST hosts, but if only EQ is involved, EQ APO is a nice all in one solution for the mighty price of free. it even has a convolver if you end up slightly mad(I noticed we have that in common) and want to also mimic phase and stuff to turn one IEM into another one(minus non linear stuff).

 

[SilverEars]

It would be interesting to have you A/B both these headphones. Maybe you'd think the same of the KSE1500? If you thought they sounded wildly different, now that would be interesting, because then we'd have to figure out what your ears can pick up that these mics can't - or perhaps something we're not looking at closely enough in the measurements. I still suspect our ears are quite sensitive to FR changes that might look subtle in a graph, but actually aren't - i.e., we'd easily see those differences if we zoomed in on the frequency and amplitude axes.

I'm curious and very interested.

Unfortunately, I don't have the iems to compare. If anybody can let me barrow for comparison, I can provide feedback on the sound compared to the graph. I have good trader feedback, and been involved in tours so I can be trusted for borrowing the iems.

I have in the past seen graphs that do not respresent what I heard. Hifiman RE400 and 600 were like that.

This can be just inaccurate graph, my ear difference from the rig, or something else. Interested in finding out more.

 

[csglinux]

Many thanks for the EQ suggestions @buonassi and @castleofargh. I will check these out at the weekend. What are others experiences of True-fi or other EQs to match SPL vs frequency (@dc655321 is correct - I need to be careful with my terminology). Has anybody had the experience of EQ perfectly turning one headphone into a direct clone of another?

@SilverEars - where are you located? You anywhere near LA?

 

[castleofargh]

Many thanks for the EQ suggestions @buonassi and @castleofargh. I will check these out at the weekend. What are others experiences of True-fi or other EQs to match SPL vs frequency (@dc655321 is correct - I need to be careful with my terminology). Has anybody had the experience of EQ perfectly turning one headphone into a direct clone of another?

@SilverEars - where are you located? You anywhere near LA?

perfect is a big request. I've had some sighted experience that seemed to match IEMs pretty well, but TBH how would I go about testing that properly as a listening test? IDK, so I'm not going to claim anything.
in practice, there is always the question of both the IEM and the amplifier handling a massive EQ without trouble. then there is the danger of trying to EQ big fluctuations precisely located in the trebles when we know that the slightest insertion change might shift it all.
for all those reasons, I wouldn't expect True-fi or any other such system to actually try to make identical copies. instead I expect them to compromise for each model of headphone while simply aiming toward a general target.

@RRod was pretty happy with his convolutions between 2 headphones, you can PM him and understand all the math I didn't when we discussed it.

 

[james444]

I have a deep(?) philosophical question for all you measurement experts. This particular exchange has had me scratching my head for while now:

https://www.head-fi.org/threads/rha-cl2-tuning-preferences.894595/#post-14635751

Is this really possible? Two headphones measure very similarly, yet sound nothing alike? I've seen this type of statement thrown around before - sometimes as an attempt to discredit measurements.

From my experience with IEM modding, I have to say that dynamic driver IEMs with similar FR sound always very much alike. I can't say the same when comparing similar measuring DDs and BAs though, in which case the latter usually seem to sound slightly leaner in bass and slightly edgier in treble to my ears.

I think it's also worth mentioning that similar FR in a coupler doesn't necessarily imply similar FR in your ears.

 

[buonassi]

Many thanks for the EQ suggestions @buonassi and @castleofargh. I will check these out at the weekend. What are others experiences of True-fi or other EQs to match SPL vs frequency (@dc655321 is correct - I need to be careful with my terminology). Has anybody had the experience of EQ perfectly turning one headphone into a direct clone of another?

@SilverEars - where are you located? You anywhere near LA?

Making one headphone sound like another? Well, that's a tall order as it's not just about FR ....and as other have mentioned some systems can't handle the compensation curves without introducing distortion.

I find EQ helpful for correcting minor peaks and nulls in the native response that either omit or mask detail. It's amazing what a couple dBs in the right places (above 1k) can do for tambre and naturalness while retaining the headphone's overall intended tuning .

 

[crinacle]

For the hardcore, I've finally finished fixing my old measurement files. All 1,294 data points (as of 09/03/19) are available.

https://crinacle.com/graphs/raw-data-files/

 

[malvinviriya]

[Quote of deleted post removed]

Why so much hate

Honestly I do have the same setup as @crinacle. Both his legacy setup and his current one. I happen to be able to replicate it to almost a tee since we’re both using iPhones and that platform really allows for a mobile measurement setup

Fun fact though, if you see the first page he also credited me and NMatheis for providing measurements where he can’t provide

Now, on to the recent topic, since I’ve fully replicated his setup why don’t I just post my measurements here right? For that, personally I feel it’s due to respect. The only reason I was credited for my measurements was because he found something interesting to be measured (Avara’s lineup). He asked me for measurements and I provided. This is his thread and especially considering Crinacle seems to want to go professional on this, I don’t think it would be right for me to constantly post unsolicited measurements. Discussions are fine and backing it up with your own data is cool but keep it at a limit. As I said in a post far above, it’s all about keeping respect as you’re sort of in someone else’s “home”.

If I were to provide my own measurements I can definitely do so. There’s plenty of new and hot products to be measured right here. I can definitely eclipse Crinacle in terms of the frequency of new units and whatnot. But I’ll be capitalizing on Crinacle’s thread popularity to merely bolster my online image.

Call me old schcool, but I just prefer to attract my own audience by not leeching from anyone

 

[castleofargh]

if you guys really plan to do that, give me a link to the thread because that's sort of my porn. and I'd like to suggest what we never got around to doing with Tyll when he was still doing measurements(part of it was my fault TBH as I asked right before leaving on holidays and then forgot all about it...), the idea is to have one or ideally several pairs of IEMs go around the people participating, to serve as reference for calibration. I'm thinking one BA, one dynamic, one that goes deeps, one that cannot, or something like that. also I think we need several because one of those days, the IEM will break for reasons, and after that new guys will be screwed if it is the only reference.

we could simply pick an IEM and use it as flat reference(maybe with a given insertion depth, or simply by telling people that the stuff should resonate around XkHz in the trebles). so people would just have to put the IEM in the coupler, and press calibrate on whatever app they have to create a usable compensation.
or given how many IEMs @crinacle already has measured, maybe we could use him as reference, he measures the "calibration IEMs" like he always does, and then we all calibrate to get the same response he does when we get our tour to measure them.

or any other option. when it comes to calibration, it's going to be arbitrary anyway, so why not take something with a practical benefit. @crinacle seems like a de facto reference nowadays and maybe he doesn't want to have to apply a new compensation to all his files and upload them all again. the point being that we would agree to pass around a set of IEMs that would from then on, only be used for that purpose.

if no big reviewer wants to share some of his personal treasure's trove for this, I could provide a few or my old cheapo IEMs. the one thing I ask is that someone else takes custody of them and deals with sending and getting them back after the first tour, for new guys who wish to join in. I have 2 reasons for that:
1/ I don't want to do it. ^_^
2/ the people delivering stuff at my house basically don't. they usually refuse to go down my street because it's a dead end and they are such skilled drivers that they're afraid of having to go backward for maybe 80 meters... so any delivery turns out to be a huge PITA for me, I often don't even know that something is coming as they won't even come to give me the delivery notice. to the point that maybe a 5th of the stuff I order online, never reaches my house and I often just cancel the order after a while(it's kind of mental TBH).

anyway, this is how you hijack a topic!

 

[csglinux]

if you guys really plan to do that, give me a link to the thread because that's sort of my porn. and I'd like to suggest what we never got around to doing with Tyll when he was still doing measurements(part of it was my fault TBH as I asked right before leaving on holidays and then forgot all about it...), the idea is to have one or ideally several pairs of IEMs go around the people participating, to serve as reference for calibration. I'm thinking one BA, one dynamic, one that goes deeps, one that cannot, or something like that. also I think we need several because one of those days, the IEM will break for reasons, and after that new guys will be screwed if it is the only reference.

we could simply pick an IEM and use it as flat reference(maybe with a given insertion depth, or simply by telling people that the stuff should resonate around XkHz in the trebles). so people would just have to put the IEM in the coupler, and press calibrate on whatever app they have to create a usable compensation.
or given how many IEMs @crinacle already has measured, maybe we could use him as reference, he measures the "calibration IEMs" like he always does, and then we all calibrate to get the same response he does when we get our tour to measure them.

or any other option. when it comes to calibration, it's going to be arbitrary anyway, so why not take something with a practical benefit. @crinacle seems like a de facto reference nowadays and maybe he doesn't want to have to apply a new compensation to all his files and upload them all again. the point being that we would agree to pass around a set of IEMs that would from then on, only be used for that purpose.

if no big reviewer wants to share some of his personal treasure's trove for this, I could provide a few or my old cheapo IEMs. the one thing I ask is that someone else takes custody of them and deals with sending and getting them back after the first tour, for new guys who wish to join in. I have 2 reasons for that:
1/ I don't want to do it. ^_^
2/ the people delivering stuff at my house basically don't. they usually refuse to go down my street because it's a dead end and they are such skilled drivers that they're afraid of having to go backward for maybe 80 meters... so any delivery turns out to be a huge PITA for me, I often don't even know that something is coming as they won't even come to give me the delivery notice. to the point that maybe a 5th of the stuff I order online, never reaches my house and I often just cancel the order after a while(it's kind of mental TBH).

anyway, this is how you hijack a topic!

Great idea buddy! I'd suggest an Etymotic with foam tips would be a good starting point for a non-ported IEM. Maybe Zero Audio Carbo Tenore as a ported reference (as it's also relatively affordable)?

I strongly agree with this point:

1/ I don't want to do it. ^_^

because I don't want to do it either . ^_^

It is going to require a bit of effort from the OP, plus a willingness to pay for some shipping (and, if we're ever going to use USPS, insurance too). Shipping anything internationally likely means accepting some risk. Maybe all those willing to join the project donate $50 via PayPal/CashApp/etc, to the OP in order to be on the mailing list for the reference IEMs? I'd be willing to do that. It wouldn't be fair to expect the OP to cover every outbound shipping cost.

I'd suggest:

Foam tips (the same tips could ship with the reference IEMs).
1 non-ported IEM.
1 ported IEM.
Contributors would be free to contribute their own measurements for whatever IEM they like once they've submitted 711 (or 711-corrected) raw/uncompensated measurements for the above 2 IEMS (left and right channels separately on both).

I would start with just amplitude vs frequency results. That's by far the most important metric anyway (and in my experience, THD is much harder to measure accurately with the more affordable couplers and sound cards).

I like the name "The Measurements Thread". It's nicely minimalist and could cover all sorts of additional types of measurement in the future.

 

[Otto Motor]

I started the long overdue thread: General IEM Measurements Discussions

https://www.head-fi.org/threads/general-iem-measurements-discussions.903455/

The thread is dedicated to frequency responses (and other) measurements of earphones by any Head-Fi member, using any technique and/or software! It also serves the discussion of rig setup and equipment such as couplers, microphones etc. It further encourages the discussion and exchange of compensation spreadsheets. And it promotes discussion of the tricky correlation of measurements and sound quality. Last but not least, Head-Fiers can show the visualizations of their modding steps and results as well as production flaws (e.g. miswiring and channel imbalances).

Go crazy folks!

P.S. I cleaned out my recent stuff found offensive by some and hope this solves all problems.

 

[CoiL]

@crinacle , can You PLEASE measure Moondrop Kanas Pro when You have opportunity?

 

[guicnovaes]

@crinacle Would be great see the HiFi Boy OS V3 measured by you. Can you do that? Thanks.