Frequency response is almost everything imo

[sovran]

this thread might trigger some here but i will write it anyways so maybe you think twice about buying expensive iems with BIG unrealistically expectations beforehand.
so as a guy who used studio monitors and speakers for music listening for years, i decided to buy a pair of expensive new iems to see whats the catch about them since i started following the reviews and news about them mostly the chifi ones. mind the best iem or earbud i had was galaxy buds2 pro ( bought it on sale for 110$ here) and liked the tonality of it but wanted something with better “technical performance” so i started researching, asking users here and auto EQ ing my old outdated iems or my tws to different trendy iems to see which i will like more.
i tried auto equing so many iems, from totls like mm2, ierz1, u12t, prestige,maestro se to midrange iems like variations, oracles, meteors, sa6 series etc… i found oracle mk2 to be the most clean/bright sounding and close to the buds 2 pro actually so i bought it with a fiio k7. but as i recieved it i was in shock, they sounded amazing but they were 90% the same as my auto EQed old cheap 50$ iems or even the buds 2 pros in stock form( becauae they graph the same with some difference in treble area) but sure the technicals would be better, but i wish it was.

btw these things look absolutely gorgeous and stunning, so pretty irl

i was confused when i found that it sounded all in my head and had 0 real soundstage effects like speakers or monitors. it didn’t have a center image like speakers and everything played inside my head just like my old earbuds and cheap outdated iems. don’t get me wrong, they sounded 50 times better than my old non EQ ed iems ( resolution, bass or treble quality, imaging ) but with EQ, they sounded so similar its scary. if i had not tried auto EQ before, i would just be amazed that how good they sound as a pair of iems and would keep buying even more expensive sets.
so i started to research for days and months on youtube or read articles about sound and finally landed on other forums like audio science reviews where they suggested the fact that 90% of the sound of iems is just FR due to iems being being minimum phase devices and the rest is distortion( ideal crossover engineering), linearity or correct phasing. and soundstage, detail, technical performance are all visible in graphs and after a / b ing for days. they are correct imo.
auto EQ ing my galaxy buds to other iems changes EVERYTHING about them drastically, from transient speed or sharpness, or the perceived soundstage or imaging, or the resolution of the sound.
like u12ts were warm, had good depth to the sound but not as clear as other iems like variations. or super nova sounded warmer and tamer than oracle mk2 due to its more controlled 3k and treble and sounded analogy and warm but with more closed in feeling sound and less perceived sound height. iems with boosted 3k to 20 k areas had very great imaging compared to iems with tamer ear gain and treble. or boosting 200 hz to 500hz create a warmth that helps with drums or bass greatly but makes the replay lower res. aslo iems with flat midrange and mid bass gave that spacious sound that warmer iems can’t recreate but at the cost of mid bass loss.

the only problem for me is the treble quality on cheaper iems is not good, they either clip when EQING or don’t sound refined like the oracles treble and thats why you can’t really eq treble perfectly( also ear shape and coupler differences affect how iems graph). the oracles have great treble, they are bright and sparkly but never hurt my ears like the buds 2 does in higher volumes and thats why i keep it (this might be that i can’t get a deep fit with the buds 2s like i do with the oracles due to their design and that effects the harshness of the treble)…
i will keep the oracles because how much i like its treble compared to everything i own but 500$ more compared to my tws just to have better treble is a bit too much if you ask me. or maybe im too broke

so the conclusion, this might be silly but imo you can have almost totl experience with poweramp app that you can use it to import auto EQ profiles to it+ an smartphone+ some tws buds like galaxy buds that i use( you can use literally anything that has a graph on squig link). you can have 20 different iems on your hands everywhere you go with 20 different EQ profiles on poweramp app. when you can have totl tuning that is 90% of the experience and doesn’t need dac, amps or cables, there is no logical excuse to buy a dap, or dac with a 1k iem for example where you can spend the money on top speakers unless you earn 10k+ a month and want to support well tuned iems and engineers without stealing their tuning ieas with auto EQ or that refined treble area is worth alot for you.

[darkgod5]

I have a counterpoint to frequency being everything and all intangibles and they are a couple songs:

G.O.A.T. - YouTube Music

This song will ONLY have enough “texture” in terms of the distortion of the instruments with a single-DD like the Simgot EA500.

You can EQ a multi-BA IEM to whatever frequency response you want but it will sound “muted” or “veiled” in comparison.

Sun//Eater - YouTube Music

This song will ONLY have enough “resolution” letting you able to hear enough to transcribe the song yourself without needing to slow it down with a multi-BA IEM.

You can EQ a single-DD to whatever frequency response you want but it will sound “mushy” or “low-fi” in comparison.

This thread might trigger some here but, simply put, if you can’t hear the difference between different driver setups EQ’d to the same FR then your ears are not as good as you think they are.

Edit: of course since this is taking driver quality in consideration, this is SOLELY talking about IEMs under $500 at the time of writing. I’m sure drivers and so IEMs will improve in quality and drop in price as time goes on.

[sovran]

idk, i can turn my 10 years old older cheap single dd driver iems that sound unlistenable and absolutely horrendous as clean ,sharp or sharper sounding in the transients and note attack compared to my oracle mk2s in lower volumes( and that attributes to the dip of oracles in the 1.5 to 3 k region i assume), and the buds 2 pros i have are 2dd designs and respond well with eq and without eq, due to the close fr to the oracles, if i a/b the oracles and buds together, i can only tell the difference in some song due to the buds having a 5 to 6k peak ( tss sss sounds grinding ears)and oracles having better treble extension but the difference in most songs( around 75 to 80 db) were so minimal that i could not believe that im comparing a wireless mainstream buds to a supposedly higher tier iem connected with a thicc wire into a 1 kilogram dac/amp. and i did ab them for hours and hours in the last month or so.
of course to manage a good fr with MINIMAL distortion or NO clipping in high volumes, you need good drivers and sometimes more than 1 driver specially for treble, but it seems that my 120$ tws buds have no clipping issues even in high volumes when i use auto eq on it.
and no, i don’t think my ears are bad because i can hear up to 20k still as a 23 years old and i can even hear an old crt TV being turned on from 2 rooms away( they produce a nasty 16k high frequency hiss that drives me to suicide) also i scored all 10 db and lower in hearing test even in higher frequencies i did 6 months ago before washing my ears.
but even what if my ears are not golden and im actually deaf?

this is the review of technicalities of truthear zero and zero reds, 50$ even cheaper than the buds.

Truthear x Crinacle Zero:RED IEM Review

TRUTHEAR x Crinacle Zero IEM Review

the tia trio that is praised to hell and back yet gets technically out performed by 50$ truthear zeros

64 Audio tia Trió IEM Review

this 50$ iem is even objectively better engineered than u12t, so technically the transducer quality of a 2dd 50$ iem is better than a 2000$ iem and its measured objectively for us to see.

64 Audio U12t Review (IEM)

50$ iems nowadays have better driver quality than 2000$ totls that have 12 drivers shoved in them.and good enough distortion to be inaudible even in higher reasonable volumes.
or even the monarch mk2, a 1000$ iem and probably the most praised and hyped iem has out of phase polarity,a major flaw in wiring that makes the sound of transducers heady and decreases clarity for the soundstage and tweaking unusual imaging( here its iem so no soundstage here)and yet this iem is praised by everyone for it being super technical despite its drivers being wired backwards . ( this won’t show in graph though)

https://reference-audio-analyzer.pro/en/report/hp/theieaudio-monarch-mk2.php#gsc.tab=0

[darkgod5]

@sovran

( this won’t show in graph though)

You essentially just refuted your entire argument.

[darkgod5]

I will just add though that my opinion is that in the future graphs will tell everything (not talking solely about an FR graph) because people seriously underestimate our hardware technology and at the end of the day, IEMs and even audiophile hardware is such a niche hobby that desperately needs technological testing advancement.

But right now it’s too easy to refute frequency response being “everything” as all one needs to do is EQ one IEM to another and listen to the same song. Heck, you can even listen to a song with low DR through a powerful desktop amp if you believe distortion at over 100 dB to contribute to anything of value.

[sovran]

@darkgod5

You essentially just refuted your entire argument.

refuted witch argument? i mentioned it here and it never refute anything i said above.

also if you read my argument completely, i never said that graphs are 100%.

[darkgod5]

@sovran

also if you read my argument completely, i never said that graphs are 100%.

Ahhhh fair point. My mistake there then.

Then I think we’re in agreement that graphs are foremostly already very useful with massive future potential to get to describing 100%.

[sovran]

@darkgod5

IEMs and even audiophile hardware is such a niche hobby that desperately needs technological testing advancement.

not really when you see a mainstream galaxy buds from samsung has harman adjusted tuning and graphs better compared to 95% of “niche” iems.
if they were manufactured by the chinese, they would charge you 10k a pair because it works “magically wirelessly” and has a “audiophile” tier graph and rich would buy it and praise its clarity and magic and you had to stay in line for 1 year to get a pair but they get mass produced in millions so you can have it as cheap as 200$ and they have active noise cancelling, transparency modes and touch controls on top of comparable sound quality to top tier chinese iems.

[sovran]

@darkgod5

Then I think we’re in agreement that graphs are foremostly already very useful with massive future potential to get to describing 100%.

no, i think everything shows in graphs, but not only fr but in in magnitude response graphs, frequency response graphs, and total harmonic distortion measurements, and polarity tests at least after my experience with the 600$ iem i feel that way.

[SoundEater]

And if they were made by Americans they would charge you 100,000 for the magic sound!

[sovran]

64 audio tia distortion tws

[SoundEater]

Not just 64audio. There are many companies from America and some other countries that overcharge! The results they achieve are comparable!

[Erokh]

I tested EQing a planar to a hybrid or DD FR I own, and the results on my part were veeeeery different from the real thing.

I think that eqing can bring a huge improvements and is real fun, but I can’t trust it completely.

To me, AutoEQ as you did it can’t be trusted fully, because:

• you don’t know the real FR of YOUR set of samsung buds
• you don’t know the result the EQ has on your buds

I beleive you obtained a sound that is close to the Oracle MKII, and I believe you enjoy it very much; but I can’t follow you further than that: the deductions you make about FR from this experiment are only assumptions at this point.

The thing is though, you make me want to play with autoEq again

[llysender]

I am curous about transient response being affected by normal eq as Ive never heard changing frequency on a static all encompassing basis actually affect slam and decay meanfully.

Sure there are time domain dynamic eq to say cut the subbass off early duing drum hits to give the illusion of “faster” transients but say the moondrop SSR with its heavy sustain before decay no matter how you tweek it it wont get the speed of the final A4000 and even then you still loose alot of “detail” the software is actively cutting out the subbass data which on the A4000 is interpreted by the brain as the after hit reverb of the skin.

So yeah even with dynamic EQ I find that its much better to add data to the song and make it wetter with a fast iem then try to get a slow iem to sound fast.

Then there is SLAM which cant be EQed in. If somehow you are able to defy nature and have the magnet magically change in strength along with the diaphragm become lighter/heavier as needed I’ll be impressed.

[llysender]

I take offence with Sun//Eater sounding mushy and low-fi on single DD iems. Even something as “cheap” as the moondrop Lan is able to handle the speed and texture. Using the Tri Starshine and Triangle Comete as references.

Well although I wonder how much is due to the source being able to grip the DD.

[Mufflertape]

I’m just ready for the nuralink to beam the audio data directly to your brain so it is a perfect signal and then its just down to your brain misfiring not interpreting the data correctly. Lol I kid as that is nightmare fuel on the other implications, but at least will help you think of how it will work. Until we get there this seems more on the petty side. Make sound go good just dont stress on it.

[sovran]

@Erokh

I tested EQing a planar to a hybrid or DD FR I own, and the results on my part were veeeeery different from the real thing.

I think that eqing can bring a huge improvements and is real fun, but I can’t trust it completely.

To me, AutoEQ as you did it can’t be trusted fully, because:

• you don’t know the real FR of YOUR set of samsung buds
• you don’t know the result the EQ has on your buds

I beleive you obtained a sound that is close to the Oracle MKII, and I believe you enjoy it very much; but I can’t follow you further than that: the deductions you make about FR from this experiment are only assumptions at this point.

The thing is though, you make me want to play with autoEq again

of course you can’t 100% EQ one iem to other iem because

1. every couple measures slightly different so the measures fr on other couplers might differes slightly from yours
2. every ear canal has different shape so it slightly affects the higher frequencies, but there is an average shape to it that is close to everyone
   https://www.audiosciencereview.com/forum/index.php?attachments/preferred-response-white-paper-final-pdf.207412/
3. iems have slight variance though i saw thieaudio had good channel matching and minimal unit variation unlike kz or other cheaper brands

but i think you can get close to it if you have 2 iems with yourself and a/b back and forth to have close EQ results or measure them both and eq one to other thats unrealistic because why eq one to other if you have both?
if you don’t have it then you expect to get 80% to 90% close to it and its just depends on you luck.

[sovran]

@llysender

I am curous about transient response being affected by normal eq as Ive never heard changing frequency on a static all encompassing basis actually affect slam and decay meanfully.

Sure there are time domain dynamic eq to say cut the subbass off early duing drum hits to give the illusion of “faster” transients but say the moondrop SSR with its heavy sustain before decay no matter how you tweek it it wont get the speed of the final A4000 and even then you still loose alot of “detail” the software is actively cutting out the subbass data which on the A4000 is interpreted by the brain as the after hit reverb of the skin.

So yeah even with dynamic EQ I find that its much better to add data to the song and make it wetter with a fast iem then try to get a slow iem to sound fast.

Then there is SLAM which cant be EQed in. If somehow you are able to defy nature and have the magnet magically change in strength along with the diaphragm become lighter/heavier as needed I’ll be impressed.

you can see it on the fr
if you play with the 20 to 200hz area
or 200 to 1k
or 1k to 3k
or 3k to 6k
and 6k to 20k

you get various transient speed, attack, decay.
you can get the drum snares to sound boomy and full that has the impression of reverb and full ( boosted midbass)
or get the drums sound cut of or plastick( no midbass and boosted upper midtange)
or add that sharp edge to them to have them feel faster and crisp hitting by boosting higher frequencies and taming lower fequencies
or you can get them dull sounding by cutting off both ends

from the graph, if you play the same snare sound file on these 2 iems
the ssr would sound plasticky, hollow, bright and slightly muted in the lower freuencies
on the other hand you get a full, bity snare from the final audio but at the cost of fast songs becoming muddy and loss of resolution
you can reverb the snare if you have a boosted subbass and midbass in my experience
so my guess is that the ssri has high distortion levels so by boosting that area, you won’t get the ideal replay you want since you have to boost the subbass quite substantially to match the a4000 and thats alot of extra power the driver has to handle.

[llysender]

Interesting how you can assume so wrongly bassed off graphs,

SSR is warm with thicc midbass pressence slightly slow with a fun lower mids focused but darkish treble because of the boosted mids and simply isnt detailed at all and esp if you reduce the the 2k to 3k area to make it less mid forward you get to hear the smeary treble in all its gorey. No amount of snake oil dac/amplifier can make the SSR sound detailed but thats part of the fun of it because the midbass hits like a one note dumptruck and I would say its more of a party/edm type of iem. The last thing I would call it is plasticly, hollow and bright. Infact I would say SSR is darkish.

A4000 on the other hand dispite having that level of boosted bass because eveything dcays so fast along with the boosted high treble 8k onwards tends to sound brighted and very dry. Personally I eqed the A4000 to SSR because I understand the driver power handling charactoristics on both and that A4000 is able to take more abuse without distorting which while making it sound slightly warmer doesnt take away from the A4000 still being able to handle songs like Sun//Eater with no difficulty.

If plasticy means dry then yes A4000 is plasticy but considering I listen to speedcore on the A4000 its hard to imagne the A4000 being muddy. Although I will agree with the loss of resolution but thats because Im anal and am comparing it to the Tri Starshine and Effect Gaea which both are able to handle dynamic range well and reproduce low level sounds well while the A4000 cannot due to driver limitations.

Thank you for listening to my rant on the SSR.

[sovran]

@llysender

darkish treble

the treble in ssr is literally boosted to the point off hearing loss

@llysender

I would say its more of a party/edm type of iem

bro it literally has rolled of subbass, and for edm you need boosted subbass

there is no balance here, any edm song will literally pierce your brain. this tuning is a clarity based and meant for acoustic songs mainly strings imo.