Headphone journey

[solderdude]

Well you don't hear more bass nor does it become bassy or flabby.
Just more correct.
When listening to piano for instance you can hear the 'mechanical action' better for instance.
It improves realism more. Bass is still tight and sound stage is still good. Just not as 'hyper-detailed' just realistic detail.

It is easy to EQ headphones based on the before and after plots to give it a try.
The subbass boost is less than that of Sonarworks.

 

[Sal1950]

At one time headphones were an island of sanity in the dysfunctional world of audio and companies like Sennheiser, AKG, Beyerdynamic and Sony made worthy well engineered headphones that avoided all the audiophile insanity for the most part. The last decade or so has seen a proliferation of high priced headphones along with all that goes with audiophile madness such as silly cables, ridiculous amplifiers, flowery subjective reviews in websites and magazines to push the gullible into parting with $$$$$$$$$`s etc. I just find it a little sad.

I keep a older pair of Grado SR80s on the nightstand next to the bed and last night I was listening to Phil Collins - Face Value. It hit how amazing they sounded being driven with an old cheap Nextbook tablet. Not sure where I'm really going with this post but just thinking about that point in the cost curve where things get totally out of hand in the performance vs $. No these $99 Grado's don't match my Senn HD650's at the freq extremes or inner detail, but the rig does supply most everything that is important for a very enjoyable listening experience. Sadly it seems Grado has lost favor in the headphone crowd lately?

 

[solderdude]

I guess I have no idea how to EQ a headphones. It sounds complicated haha

The trick is NOT to EQ it opposite direct measurements as no measurement will be equal to what you hear.
I measure what the driver does and only apply some general (not exact) EQ to compensate what the headphone needs overall.
So no sharp peaks or dips are 'corrected' based on a measurement.

There are many ways to EQ and some will improve it some may even make certain aspects worse. One can base EQ on measurements from their favorite site or use ears only or base it on other tricks.

 

[MRC01]

The one that a lot of people don't like because they don't EQ it (properly) ... the HD800.
I don't care for it in stock condition so I get why some don't like it (in stock condition).
Bought it secondhand.

The HD-800 sounds fine with a notch filter, parametric -6 dB at 6 kHz, Q=2. Otherwise it's bright and fatiguing. The lowest bass octave is a bit attenuated, but if you boost it gets a bit wooly. People who like that bottom bass octave should try a good planar. Incredibly taut and clean, linear, extending to subsonic.
Good tight bass isn't just for pop music. The lowest harp strings have a fundamental around 25 Hz.

 

[3vilgenius]

Well you don't hear more bass nor does it become bassy or flabby.
Just more correct.
When listening to piano for instance you can hear the 'mechanical action' better for instance.
It improves realism more. Bass is still tight and sound stage is still good. Just not as 'hyper-detailed' just realistic detail.

It is easy to EQ headphones based on the before and after plots to give it a try.
The subbass boost is less than that of Sonarworks.

Have you not used the Kameleon amplifier to achieve this FR curve with the HD 800?
https://diyaudioheaven.files.wordpress.com/2013/12/tp-effect-on-kameleon-hd8001.png?w=1228

That is SUPER far reaching Sub base according to this FR. Doesn't that mean there HAS to be more base? Or is this not the EQ you've been using? If not, have you heard this FR achieved with the HD 800 using the kameleon before?

 

[solderdude]

Yes, with the Kameleon + 1 ply of TP over the driver to lower the 11kHz peak a bit.
There is more subbass.. not flabby at all. Just more 'body' when it is in the recording and corrected for 'comfortable long-haul listening sessions.
At loud demonstration levels which one can only sustain for say half a song or so the bass becomes a bit elevated but not annoyingly so.
Bass is usually in the 60Hz to 200Hz region. The bass is 'raised' a bit in the O-W kind of way.. specifically to compensate studio recordings mastered between 80 and 85dB so the headphone sounds 'correct' in the bass at levels between 65 and 70dB.

 

[3vilgenius]

The bass is 'raised' a bit in the O-W kind of way.. specifically to compensate studio recordings mastered between 80 and 85dB so the headphone sounds 'correct' in the bass at levels between 65 and 70dB.

What does the O-W way mean? I'm new to the lingo. Have you heard tracks with bass that goes into the 30hz region? For example the bass line in Kendrick Lamar's track "Untitled 02". Does it have any thump when the initial bass beat comes in and sustain decent rumble after? This is the bass that most people on this form don't really hear on their music but it's so essential to the music that I listen too. I don't even consider myself a bass head I just want it to be represented properly even in ranges as low as this. And if I can EQ and HD 800 to just have the presence I described that's and end game can for me right there with the Kameleon and TP mod

 

[solderdude]

Olive-Welti (also known as Harman curve), just google it.

Not heard Kendrick Lamar.
Just more realistic, not more bassy with well made recordings at pleasant listening levels. Most people (incl. me) find the stock HD800 to be lean on the bass, slightly 'anemic' but can still hear subbass extension, just not at the proper 'level' at lower listening levels.
It is only corrected for this.. more real at lower listening levels, not boosted.
When one overdoes the EQ by just 1 or 2 dB's the bass becomes fat and wooly and when it doesn't follow the correct curve it sounds wrong to me as well. There is only a very small 'correct' window which is recording and listening level dependent as well.
regular pop music usually has 'squashed' bass with heavy compression and this will still sound the same but perhaps with some more 'body' to it.
No teeth rattling.

 

[daftcombo]

I looked up the EQ corrections on the DIY audioheaven website and holy cow, the sub base on those things with your eq's are straight up unbelievable. Like I straight up can't believe what I'm seeing haha. So tell me... what's it like? I listen to a lot of rap and R&B and I've realized that the majority of audiophiles straight up don't listen to the music that I do, so I can never trust anyone's opinion on base. But those measurements for the EQ'd 800's and the 650's for that matter... is it as good as the FR shows? Also after EQ have you sacrificed any of the sound stage bigness that make the 800's what they are?

Perhaps the searcher got some of them from DIYheaven actually, I don't know. I highly doubt he tested and made all those measurements himself since he is more interested in binaural sound and the EQ is only a way to help achieving it.

 

[daftcombo]

I looked up the EQ corrections on the DIY audioheaven website and holy cow, the sub base on those things with your eq's are straight up unbelievable. Like I straight up can't believe what I'm seeing haha. So tell me... what's it like? I listen to a lot of rap and R&B and I've realized that the majority of audiophiles straight up don't listen to the music that I do, so I can never trust anyone's opinion on base. But those measurements for the EQ'd 800's and the 650's for that matter... is it as good as the FR shows? Also after EQ have you sacrificed any of the sound stage bigness that make the 800's what they are?

Once properly EQ'd, all headphones should sound the same !
Sure, you'll have bass a-plenty, provided the phones are well sealed on your ears. The soundstage will be better as well - at least it more realistic.
Not flat headphones or speakers act as if there were an effect box in the chain. You might like the sounnd of some songs better like that. But if you want to listen as it was mastered, flat response is desirable.

 

[MRC01]

Once properly EQ'd, all headphones should sound the same ! ...

Did you forget a smiley?
There is more to what we hear than FR. Distortion, impulse, phase, CSD, etc. Headphones sound different even if you EQ them to the same curve.

 

[3vilgenius]

Did you forget a smiley?
There is more to what we hear than FR. Distortion, impulse, phase, CSD, etc. Headphones sound different even if you EQ them to the same curve.

I would hope so cuz otherwise I'd have an audiophile crises. What would be the full list of "unmeasurables" that headphones all have? I would like to know because (to kinda veer back on topic of OP's post) I believe its these things that are left as we continue headphone our journey. I love a flat response because I want true to music sound, but that being said if we do indeed EQ headphones to sound somewhat similar in FR, there MUST be X thing that still separates how they literally sound to us as a whole.

 

[MRC01]

These other aspects of sound are measurable. They amount to distortion and time domain performance. You can EQ it flat yet that can make it more distorted, bloated and worsen ringing. Sometimes the difference between medicine and poison is the dosage; too much of the cure (EQ) can be worse than the disease.
I too prefer a flat response yet with headphones the definition isn't as simple as it is with speakers because we each have an individual and different HRTF. What is 'flat' depends on the listener.

 

[daftcombo]

Agreed with all BUT IMHO frequency response is the most important parameter and perhaps the only audible at low or mid volume.

 

[solderdude]

Once properly EQ'd, all headphones should sound the same !

No they don't. One can EQ in the electrical sense but has NO control over acoustics, resonances, break-up in driver membranes, driver angles and more. You can add upper treble and it may measure better in FR but not sound any better.

And ... measure the exact same headphones on 4 different rigs and you get 4 different plots which when used to EQ 'exactly' will yield 4 different sound signatures, some worse sounding than others as certain dips and peaks aren't really there or in other amplitudes and get EQ'ed incorrectly.

For this reason I just look at the 'general' errors one sees in all of these different measurements and make a 'general' (= NOT exact) EQ that corrects what most likely needs correcting and base my Kameleon filters on my own measurements (how flawed they may be in other eyes) AND use my ears + reference headphones to compare against.

Not all headphones react well to EQ b.t.w. Only the better ones that just need a nudge here and there are eligible.

FR is a very important factor to get the tonal balance correct. This will make the headphone sound best (most realistic) with well made recordings but expose loudness war-ed and manipulated recordings.
Such popular music recordings may well sound much better on bassy headphones with a huge 3kHz dip and rolled off treble. Well made recordings will sound like poop on those headphones.

 

[M00ndancer]

I run EQ on my Tin Audio T2s on the phone, making them suit my preferences. Sound really really good.

 

[daftcombo]

No they don't. One can EQ in the electrical sense but has NO control over acoustics, resonances, break-up in driver membranes, driver angles and more. You can add upper treble and it may measure better in FR but not sound any better.

And ... measure the exact same headphones on 4 different rigs and you get 4 different plots which when used to EQ 'exactly' will yield 4 different sound signatures, some worse sounding than others as certain dips and peaks aren't really there or in other amplitudes and get EQ'ed incorrectly.

For this reason I just look at the 'general' errors one sees in all of these different measurements and make a 'general' (= NOT exact) EQ that corrects what most likely needs correcting and base my Kameleon filters on my own measurements (how flawed they may be in other eyes) AND use my ears + reference headphones to compare against.

Not all headphones react well to EQ b.t.w. Only the better ones that just need a nudge here and there are eligible.

FR is a very important factor to get the tonal balance correct. This will make the headphone sound best (most realistic) with well made recordings but expose loudness war-ed and manipulated recordings.
Such popular music recordings may well sound much better on bassy headphones with a huge 3kHz dip and rolled off treble. Well made recordings will sound like poop on those headphones.

You are more specialist than me, yet basically you do as I say: EQ and compare to a reference until you almost reach it.
Among the EQs I linked are several Average files made out of different measurements of the same product. No big Q involved I think.

I agree with all you said btw. From head to head the FR changes, like speakers with rooms.

Out of curiosity what are your reference phones? Are they EQed?

 

[solderdude]

EQ'ed HD650 for tonal accuracy and for spatial cues and distortion an EQ'ed HD800.
Both are not modified.
I am not particular a Sennheiser fan but these happen to be easiest to 'compensate' well.

Some of the convolved EQ's that exist are directly based on measurements and a program converts that to EQ. Sometimes with high Q's.
That's what I was referring to. This may seem 'ideal' and when one measures the result on the same rig it will measure perfectly.
Measure it on another rig or listen to it and it won't be as flat as it measures.

So indeed, a gentle but well executed general 'correction' may well improve fidelity in a lot of headphones.
Regardless if it is done by ear or based on measurements as long as the 'correct' reference is used.
That last bit is usually the point of discussion...

 

[daftcombo]

Some of the convolved EQ's that exist are directly based on measurements and a program converts that to EQ. Sometimes with high Q's.
That's what I was referring to. This may seem 'ideal' and when one measures the result on the same rig it will measure perfectly.
Measure it on another rig or listen to it and it won't be as flat as it measures.

Is it because EQ with high Q (>1) add a lot of distortion?

 

[solderdude]

It adds ringing in the band it works (audible band) but the reason why not to use sharp 'correction' filters is that narrow dips and peaks may not be present when measured with a HATS or other measurement gear and could be a measurement artifact instead of reality.
You could be compensating something that does not need compensating and would 'f' things up.
Sometimes an FR plot shows a null but when looking at the CSD you find that it just takes some time to 'swing' in the correct amplitude.
In those cases an FR based on white noise and averaged over a few minutes could give better clues than a swept sine.