Good somewhat neutral mixing headphones with good bass extension for max. 350€?

[FTB]

you dont know what hrtf is, and those who mix well on planars can also mix well on 10 dollar DD iems if they have enough time.

I have a Hifiman HE1000 V2 Stealth and everything translates better on my monitors that cost 1/4th the price.

You can disagree with your anatomy all you want but the fact is headphones arent the best for mixing.

You say your monitors sound better but did you see how the HE1000 stealth measure ? (badly) :

Dropbox

www.dropbox.com

They are not a good reference point as headphones, comparing them to monitors is pointless.

 

[FireEmblem]

HD650 does not have less bass extension, just more midbass hump (warmth) around 140Hz which the JT7 lacks.

Good catch, I overlooked that the graph for JT7 ends 10Hz lower so that's why I thought it has less bass. However, if the lowend is pretty much similar, it would rather be a step to the side than an upgrade for monitoring low end in headphones.

I'm gonna download the free trial from Sonarworks and see what the H650 lowend will sound like with it, because I think the 650 is not far from the sweet spot.

Is boosting the lows something that reduces the lifespan of headphones?

 

[Jiraya369]

You say your monitors sound better but did you see how the HE1000 stealth measure ? (badly) :

Dropbox

www.dropbox.com

They are not a good reference point as headphones, comparing them to monitors is pointless.

TL;DR - You're wrong.

the full post :

they measure badly? woah man i dont really care. I've EQ'd it to my own preferences after doing an autoeq to harman. I now love it more than any other headphone I'd say, unless that headphone has the exact same tuning. That's just how my HRTF is, this eq sounds closest to my Adam D3V with a slight reduction eq around 8khz and room modes fixed. Even then i dont like it much.

Even Amir liked the Hifiman Arya, which had worse tuning and even worse distortion than this pair. Yet he gave it his highest recommendation with EQ. That's how good the audio reproduction capabilities are of hifiman headphones.

There's no good reference point for comparing headphones to monitors. Unless you're a mixing engineer or something I don't really know what to say mate, you're just saying things randomly without trying them. Mixing is a skill and your gear is what allows you to perform the task. Yeah, some people might like wearing sandals for climbing mountains and you CAN do it, but it's not the best way of doing that task is it? That's been my ultimate point.

Go ahead, get a neutral studio monitor and a "neutral" pair of headphones. Without getting used to them, make a mix on both of them and see which translates better. Forget tonality/eq'ing, even reverb mixing is easier with monitors and translates exceptionally well to everything. I'm no pro mixing engineer, just some guy who mixes things and plays around with stuff as a hobby and even then monitors just rule.

Maybe you have golden ears, maybe you are the perfect specimen for headphone mixing. That I am not, and neither are 99% of people. Sounds are best reproduced on loudspeakers and monitoring, mixing and mastering is best done on them as well. Headphones and iems are compromises.

This is my eq :

Btw keep in mind the harman target is just a preference curve and there's no set neutral target for headphones. The harman target was made with the old gras or something rigs and the best rig right now is the 5128. Think of the 5128 as the klippel of headphones. As good as it gets for headphone measurements. And there's no harman target for that, just a diffuse field one, and since it's not done by harman, will you consider this an actual reference point?

 

[solderdude]

Good catch, I overlooked that the graph for JT7 ends 10Hz lower so that's why I thought it has less bass. However, if the lowend is pretty much similar, it would rather be a step to the side than an upgrade for monitoring low end in headphones.

I'm gonna download the free trial from Sonarworks and see what the H650 lowend will sound like with it, because I think the 650 is not far from the sweet spot.

Is boosting the lows something that reduces the lifespan of headphones?

It won't.
Do note that Sonarworks boosts the bass even higher than Harman and in a different way.
Sounds impressive but can lead to bass shy recordings.
Here too... learn how the EQ'ed headphone 'translates' to a good reference.

HD650 with EQ is a great headphone.

 

[solderdude]

The harman target was made with the old gras or something rigs and the best rig right now is the 5128. Think of the 5128 as the klippel of headphones. As good as it gets for headphone measurements. And there's no harman target for that, just a diffuse field one, and since it's not done by harman, will you consider this an actual reference point?

Opinions differ on the 5128 as they also do on target curves. Dr. Olive agrees on that point too.
It certainly is not 'the Klippel NFS' for headphones.
BK5128 mostly differs in the way the simulated ear canal is shaped and the microphone is positioned which is closer to that of a human ear canal than the previous standards. Perfectly usable for research on acoustics but the older standards were also pretty good at that.

The pinna that was used for the Harman target was not even a standard pinna but something they cooked up themselves.
This does not invalidate the very extensive Harman research nor the reason why they started the research nor why they made the choices they did to determine a (their) standard.

References (test fixtures + target curves) are for research so one can get comparable results when doing research.
Product testing is another thing than research but one can use the same tools.

side note:

I am still amazed that people really believe their pinnae, ear canals, ear condition, seating for the measurement, seal, (silent) revisions, product variance, pad condition and preference actually matches that of any test fixture with a certain copy of a headphone and genuinely believe that importing some EQ will turn it into a true reference for them.
The science is clear about that too (that it isn't but that it complies to a standard).
Sure, in most cases (the 'majority' of people) it will make headphones sound 'better' to them so that's the win.
A (slightly smaller) 'minority' prefers a different 'sound'. Those are usually the 'I hate Harman sound' people for whatever reason they might have.

Regardless how well a curve fits to some standard it is bound to differ from personal perception.
The proof is the fact that people's opinions on the sound of each headphone model differs from 'the best there is' to 'it sounds like total crap'.
Below 100Hz and above 6kHz all test fixtures and curves have an increasing degree of error.
Older people may remember the heated debates about which headphone was 'best'... the HD650, DT880 or K701. Consensus was never reached.

Everyone that loves Oratory's work and actually looks at the pdf's (and not blindly downloads a curve) knows that his measurements have limitations, are not done on the industry standard Harman uses (but is close to it and also is not the BK5128) may have noticed the text on the pdf to tune certain bands to taste and even suggests which bands they are.

When a headphone sounds good and shows no major problems on (different) test fixtures it can be successfully used for mixing when one knows how it 'translates' to a reference. That's where the trick lies and the important part here is the used 'reference' (speakers in a room usually).

The same goes for music enjoyment b.t.w.
After all the recording is all we have and in order for that to sound good the reference during the recording must be good (circle of confusion).

 

[FTB]

TL;DR - You're wrong.

the full post :

they measure badly? woah man i dont really care. I've EQ'd it to my own preferences after doing an autoeq to harman. I now love it more than any other headphone I'd say, unless that headphone has the exact same tuning. That's just how my HRTF is, this eq sounds closest to my Adam D3V with a slight reduction eq around 8khz and room modes fixed. Even then i dont like it much.

Even Amir liked the Hifiman Arya, which had worse tuning and even worse distortion than this pair. Yet he gave it his highest recommendation with EQ. That's how good the audio reproduction capabilities are of hifiman headphones.

There's no good reference point for comparing headphones to monitors. Unless you're a mixing engineer or something I don't really know what to say mate, you're just saying things randomly without trying them. Mixing is a skill and your gear is what allows you to perform the task. Yeah, some people might like wearing sandals for climbing mountains and you CAN do it, but it's not the best way of doing that task is it? That's been my ultimate point.

Go ahead, get a neutral studio monitor and a "neutral" pair of headphones. Without getting used to them, make a mix on both of them and see which translates better. Forget tonality/eq'ing, even reverb mixing is easier with monitors and translates exceptionally well to everything. I'm no pro mixing engineer, just some guy who mixes things and plays around with stuff as a hobby and even then monitors just rule.

Maybe you have golden ears, maybe you are the perfect specimen for headphone mixing. That I am not, and neither are 99% of people. Sounds are best reproduced on loudspeakers and monitoring, mixing and mastering is best done on them as well. Headphones and iems are compromises.

This is my eq :
View attachment 498725

Btw keep in mind the harman target is just a preference curve and there's no set neutral target for headphones. The harman target was made with the old gras or something rigs and the best rig right now is the 5128. Think of the 5128 as the klippel of headphones. As good as it gets for headphone measurements. And there's no harman target for that, just a diffuse field one, and since it's not done by harman, will you consider this an actual reference point?

Why don't you choose headphones which sound good without EQ ? It exists.
Harman target is a preference curve obtained by objective methods, it is a new standard for headphones, better than no standard at all.

B&K 5128 is not at all the best rig, it's hype made by headphones.com and others reviewers saying "the Harman target is dead, mine is better !" :

AES 2025 Paper: New targets for the B&K 5128 GRAS 45CA-10

There is a new AES paper published by Dr. Sean Olive @Sean Olive and Dan Clark @Dan Clark on potential targets for GRAS 45CA and B&K 5128: A Preferred Headphone Target Curve Defined for the B&K 5128and GRAS 45CA-10 Test Fixtures ABSTRACT The Harman around-ear and on-ear headphone target...

www.audiosciencereview.com

My headphones, k371 without EQ (sounding neutral, yes !) are way better than my monitors in untreated room, Kali LP6 v2 (pretty linear for the price).
Here is the Kali smoothed at 1/12 :

I am not sure a delta of 20 db between 50 and 150 Hz can be tamed with acoustic treatment.

 

[Jiraya369]

Opinions differ on the 5128 as they also do on target curves. Dr. Olive agrees on that point too.
It certainly is not 'the Klippel NFS' for headphones.
BK5128 mostly differs in the way the simulated ear canal is shaped and the microphone is positioned which is closer to that of a human ear canal than the previous standards. Perfectly usable for research on acoustics but the older standards were also pretty good at that.

The pinna that was used for the Harman target was not even a standard pinna but something they cooked up themselves.
This does not invalidate the very extensive Harman research nor the reason why they started the research nor why they made the choices they did to determine a (their) standard.

References (test fixtures + target curves) are for research so one can get comparable results when doing research.
Product testing is another thing than research but one can use the same tools.

side note:

I am still amazed that people really believe their pinnae, ear canals, ear condition, seating for the measurement, seal, (silent) revisions, product variance, pad condition and preference actually matches that of any test fixture with a certain copy of a headphone and genuinely believe that importing some EQ will turn it into a true reference for them.
The science is clear about that too (that it isn't but that it complies to a standard).
Sure, in most cases (the 'majority' of people) it will make headphones sound 'better' to them so that's the win.
A (slightly smaller) 'minority' prefers a different 'sound'. Those are usually the 'I hate Harman sound' people for whatever reason they might have.

Regardless how well a curve fits to some standard it is bound to differ from personal perception.
The proof is the fact that people's opinions on the sound of each headphone model differs from 'the best there is' to 'it sounds like total crap'.
Below 100Hz and above 6kHz all test fixtures and curves have an increasing degree of error.
Older people may remember the heated debates about which headphone was 'best'... the HD650, DT880 or K701. Consensus was never reached.

Everyone that loves Oratory's work and actually looks at the pdf's (and not blindly downloads a curve) knows that his measurements have limitations, are not done on the industry standard Harman uses (but is close to it and also is not the BK5128) may have noticed the text on the pdf to tune certain bands to taste and even suggests which bands they are.

When a headphone sounds good and shows no major problems on (different) test fixtures it can be successfully used for mixing when one knows how it 'translates' to a reference. That's where the trick lies and the important part here is the used 'reference' (speakers in a room usually).

The same goes for music enjoyment b.t.w.
After all the recording is all we have and in order for that to sound good the reference during the recording must be good (circle of confusion).

yeah we seem to mostly be in agreement. There's no true reference, just something to HELP guide us to different places. And yeah, the 5128 aint no actual klippel but it's the best we got now and at least for iems it's perfect imo.

the thing is though, you can't really get a neutral pair of headphones, at least for your own head, without having a proper reference, that being an accurate studio monitor. That is what I've been saying from the start. And even then, it just won't translate well enough, you can't reliably tune a headphone to the most perfectly neutral sound by ear. And even if the tonality were right, decisions like compression/reverb are far tougher to decipher properly. It's possible, don't get me wrong, but you REALLY need to learn your gear, and at that point there's honestly no need for a proper "neutral" pair, just something that plays full range without much distortion to colour the sound. Just get a headphone that you enjoy listening to, listen to it for months, make mixes and check on other systems to learn where you make errors and correct those.

I've made mixes on my Moondrop Aria (which i experimented a LOT with in terms of eq, IEF 2020, IEF 2023, Harman, DF, i tried em all and settled on a mix of IEF 2023 and IEF 2020 with a bass boost) after a year or so of use that translated well to everything and sound nice on my monitors as well, i dont think i could've done much of a better job on my monitors but what i am sure of is that the process would've been easier. Even after knowing my gear in and out, I had to test LOTS of times and then go back and forth with other sound systems, not to mention relying on iZotope's tonal balance control 2 and using a 4.5 db per octave noise sample as reference for EVERYTHING. I can do all that work in half the time without much hassle on my monitors.

So yeah, even with my HE1K Stealth eq'd to what sounds most natural to me, I still can't get mixes as nice as on my monitors. Maybe my ears suck but i just think headphones cant be compared to monitors at ALL as a tool.

 

[Jiraya369]

Why don't you choose headphones which sound good without EQ ? It exists.
Harman target is a preference curve obtained by objective methods, it is a new standard for headphones, better than no standard at all.

B&K 5128 is not at all the best rig, it's hype made by headphones.com and others reviewers saying "the Harman target is dead, mine is better !" :

AES 2025 Paper: New targets for the B&K 5128 GRAS 45CA-10

There is a new AES paper published by Dr. Sean Olive @Sean Olive and Dan Clark @Dan Clark on potential targets for GRAS 45CA and B&K 5128: A Preferred Headphone Target Curve Defined for the B&K 5128and GRAS 45CA-10 Test Fixtures ABSTRACT The Harman around-ear and on-ear headphone target...

www.audiosciencereview.com

My headphones, k371 without EQ (sounding neutral, yes !) are way better than my monitors in untreated room, Kali LP6 v2 (pretty linear for the price).
Here is the Kali smoothed at 1/12 :
View attachment 498817
I am not sure a delta of 20 db between 50 and 150 Hz can be tamed with acoustic treatment.

dunno about you but the 5128 is anything but poor. The main thing the 5128 always succeeds (based on my ears) in is better high frequency measurements. I can't trust GRAS measurements above 6-8k but 5128 measurements line up completely with my instruments. The 5128 measurements here also agree with the GRAS ones quite well. I dont trust headphones.com at all, but I do trust crinacle.

If you think your headphones sound better than your monitors, great man, enjoy them! But for mixing you need a neutral sounding monitor. Besides, I never said you should use monitors for the low end, it's tough to tame that part. Mix everything above 100-300 hz based on monitors and make tweaks on headphones for the low end. That's the budget producer way for perfect mixes. I think I said it before but I'll say it again, speakers have perfect top end and headphones have perfect low end. We hear the mid-high end mostly so it's crucial to get that right and monitors do exactly that.

I think your issue with the LP6 can be fairly easily remedied with a sub and better positioning. I moved my seat 1 feet to the right and a lot of my low end came back, still not perfect but until 100 hz i can easily trust my monitors. It's so fun making a mix and having it translate to my phone/laptop speakers INSTANTLY

 

[FireEmblem]

It won't.
Do note that Sonarworks boosts the bass even higher than Harman and in a different way.
Sounds impressive but can lead to bass shy recordings.
Here too... learn how the EQ'ed headphone 'translates' to a good reference.

HD650 with EQ is a great headphone.

I've tried it now with Sonarworks, I think the EQ sounds great. As expected it's mainly just getting the lowend a bit up and as well the treble, and a little bit in the mid. Removes the veil and makes it sound a bit more open. Nothing dramatic though, except for the lowend.

Now I wonder, if Sonarworks would be even required if a flat target curve is the only desire. I could simply just use some free pre-EQ software that works system-wide, grab the official HD650 measurement graph and just add my own EQ that just boosts or cuts the official curves towards midline, until I'm having a flat response, no?

The Superlux HD681 also sounds really great with Sonarworks, unfortunately because of the big (5dB+) differences in the pairs shipped, you can't rely on it.

 

[solderdude]

Yep you don't really need to pay for EQ (like Sonarworks) but can get equal results using free software/plug-ins, some pre-chewed EQ and a bit of tweaking.

The measured response (should you measure it) won't be 'flat' though.
Just like speakers that measure 'flat' on axis in the open field they will measure and sound different in a room despite measuring great on NFS or anechoic.

The problem with using recorded music is that you don't really know what has been done to the recording nor is it known how it actually should sound.
This is the difference between music enjoyment (tune it the way you like it) and making (reference grade) productions. Here a reference is needed.

Best would be corrected monitor speakers in a conditioned room at the listening spot ensuring you are testing at reference levels (80-85dB average SPL).
When one knows how that reference sound 'translates' on your (EQ'ed or not) headphones you can use those to check for things in the mix or, when using monitors is not possible for whatever reason, use those to mix on if that is the only option.

Audio discussed here is mostly about sound reproduction where the source is to be respected.
From there you can go 2 routes... get it to measure great (and thus reproduce the recorded material exactly, regardless of recording quality) or 'find something that sounds good to you'. This can be well measured gear but also 'poorly measuring but great sounding' gear.
There is no right or wrong ... just preference.

 

[FTB]

View attachment 498818
dunno about you but the 5128 is anything but poor. The main thing the 5128 always succeeds (based on my ears) in is better high frequency measurements. I can't trust GRAS measurements above 6-8k but 5128 measurements line up completely with my instruments. The 5128 measurements here also agree with the GRAS ones quite well. I dont trust headphones.com at all, but I do trust crinacle.

If you think your headphones sound better than your monitors, great man, enjoy them! But for mixing you need a neutral sounding monitor. Besides, I never said you should use monitors for the low end, it's tough to tame that part. Mix everything above 100-300 hz based on monitors and make tweaks on headphones for the low end. That's the budget producer way for perfect mixes. I think I said it before but I'll say it again, speakers have perfect top end and headphones have perfect low end. We hear the mid-high end mostly so it's crucial to get that right and monitors do exactly that.

I think your issue with the LP6 can be fairly easily remedied with a sub and better positioning. I moved my seat 1 feet to the right and a lot of my low end came back, still not perfect but until 100 hz i can easily trust my monitors. It's so fun making a mix and having it translate to my phone/laptop speakers INSTANTLY

Crinacle is one of these reviewers making their own target, based on what ? One person preference ?
At least Harman research with GRAS was made with a bunch of people, different age, different gender, country, trained or untrained...

 

[FireEmblem]

The measured response (should you measure it) won't be 'flat' though.

Can you elaborate on this? Sonarworks shows me an absolute ruler-flat curve on the "simulated after" curve, except for a little boost in the lows. Or are you referring to the fact that the sound will still be altered until it reaches the brain by the ear and what's inbetween? Personally I don't bother with that. Just want to ensure that I can get my headphone as close to flat as possible so that there are less things that are not coming from the source material that alter the perception.

 

[solderdude]

I think I said it before but I'll say it again, speakers have perfect top end and headphones have perfect low end.

While that is true the headphones are missing the tactile feel of speakers in the lows. It is one of the reasons for the bass boost in the Harman target. To compensate for the lack of tactile feel which is definitely used by the brain as 'input' (just like the eyes and memory are).
The bass boost for headphones is not solely mimicking 'bass boost from flat speaker in average room'.

I agree with your sentiment... mix on speakers and one can use (suited) headphones for checking aspects.
When, for some reason, using (good compensated reference speakers) is not an option a headphone can be an alternative if one knows how a 'known good recording' is supposed to sound on that headphone.

 

[solderdude]

Can you elaborate on this? Sonarworks shows me an absolute ruler-flat curve on the "simulated after" curve, except for a little boost in the lows. Or are you referring to the fact that the sound will still be altered until it reaches the brain by the ear and what's inbetween? Personally I don't bother with that. Just want to ensure that I can get my headphone as close to flat as possible so that there are less things that are not coming from the source material that alter the perception.

The ruler flat response (the same goes for oratory) is what one gets if one measures the headphone that is measured on a specific test fixture with a (perfect) seal on that particular fixture with the correction + target curve with the suggested EQ for that particular headphone in those particular circumstances.

So it will only measure 'flat' on that fixture with that EQ to the target that is used.
You are given the impression that is how the headphone should sound but is basically only true for that specific combination.

 

[FireEmblem]

The ruler flat response (the same goes for oratory) is what one gets if one measures the headphone that is measured on a specific test fixture with a (perfect) seal on that particular fixture with the correction + target curve with the suggested EQ for that particular headphone in those particular circumstances.

So it will only measure 'flat' on that fixture with that EQ to the target that is used.
You are given the impression that is how the headphone should sound but is basically only true for that specific combination.

But it's the headphone that's measured in itself, no? So when there a boost of +6dB in the headphone at let's say 1000-1200Hz, that's objectively there. And if I don't want this to be one more factor to consider when mixing, I could tone it down by reversing it with -6dB. Independently from what will happen on my head later on. But just to get rid off all the objective alterations in the first place.

 

[solderdude]

But it's the headphone that's measured in itself, no?

That's exactly the problem.
The measurement is that of the measured headphone's response on that particular test fixture (not your head) in a specific condition (seating, seal) PLUS the correction that is needed to compensate for the alterations the test fixture makes PLUS the applied target (say more bass, tilted mids/highs to mimic speakers in a room if that is the goal) PLUS the EQ derived from that measurement system + target.

That target + correction is to be based on 'something'. That 'something' is the topic of debate with the different nerdy measurebators.

 

[FireEmblem]

That's exactly the problem.
The measurement is that of the measured headphone's response on that particular test fixture (not your head) in a specific condition (seating, seal) PLUS the correction that is needed to compensate for the alterations the test fixture makes PLUS the applied target (say more bass, tilted mids/highs to mimic speakers in a room if that is the goal) PLUS the EQ derived from that measurement system + target.

That target + correction is to be based on 'something'. That 'something' is the topic of debate with the different nerdy measurebators.

Mhmmmm. So how far from the average head does this test fixture stray? Was this ever tested with some people, are the differences minor or big?

I believe some people said monitors would avoid these problems that headphones have, because the room would flatten some of the signals or something. Is that true? But in the end the sound must go through the same ear shape..

 

[Jiraya369]

While that is true the headphones are missing the tactile feel of speakers in the lows. It is one of the reasons for the bass boost in the Harman target. To compensate for the lack of tactile feel which is definitely used by the brain as 'input' (just like the eyes and memory are).
The bass boost for headphones is not solely mimicking 'bass boost from flat speaker in average room'.

I agree with your sentiment... mix on speakers and one can use (suited) headphones for checking aspects.
When, for some reason, using (good compensated reference speakers) is not an option a headphone can be an alternative if one knows how a 'known good recording' is supposed to sound on that headphone.

no doubt about it, audibility of bass is crucial for headphones cus we dont feel em like speakers

and glad that we both had a mutual agreement bud

 

[Jiraya369]

Crinacle is one of these reviewers making their own target, based on what ? One person preference ?
At least Harman research with GRAS was made with a bunch of people, different age, different gender, country, trained or untrained...

harman = good, crin = bad

that's basically your take. if you actually looked at his target and compared it to harman you wouldnt be saying anything like that. You're just negatively biased towards him and positively biased towards harman. Im not saying harman is bad, but even if the target is not 100% to your preferences, you would be blinded into liking it due to your subjective bias, which is why blind testing is so crucial.

I don't think there is much to gain with an argument here. I've laid out all my points here, if you want you can read them, if you agree then nice, if you disagree then that's nice as well let's agree to disagree.

 

[solderdude]

Mhmmmm. So how far from the average head does this test fixture stray? Was this ever tested with some people, are the differences minor or big?

I believe some people said monitors would avoid these problems that headphones have, because the room would flatten some of the signals or something. Is that true? But in the end the sound must go through the same ear shape..

This was tested by Harman and especially the bass and upper midrange/treble showed quite a large dB range variation for personal preference.
The different 'targets' such as tilted DF or Harman also leads to different results and opinions of what sounds 'realistic' and on what average listening level.
Add to that seal issues and you have a mix.
This makes it impossible to have just one target that sounds good to everyone.
The good thing about science is that one can have different standards and measurements to that standard are comparable. People's preferences (even between sound engineers) can differ though.

Speakers in rooms can have substantial issues/problems but they differ from the problems with headphone (coupling) to the head.