I spent several weeks to develop this EQ preset for Sennheiser HD800

[LaLaLard]

Hi all.
I've had my HD800 for several years but only started to try EQ since last year. The reason is I did not want to EQ with only subjective feedback. I need something to generate instantaneous, objective feedback aside from my ears.

As we know the HD800 is famous for its infamous high frequency peaks. It is necessary to use EQ whenever I listen to HD800 - I literally could not stand the original sound signature. But I couldn't EQ it properly which lead me to listen more to my speakers.

So I eventually got my headphone test rig last year and everything becomes fun since then.

Things I use:

-RME-ADI2 as the DAC + Headphone Amp;
-Fabfilter Pro-Q for EQ;
-REW for measurement;
-Audio Hijack to reroute the audio from REW (the test signal) to Fabfilter;
-HD800, serial number 01236 (early manufacture);
-MiniDSP EARS test rig.

RME ADI-2 has some of the lowest distortion in the industry so I'm pretty sure what I got out from my HD800 is going to be close to what you hear(not the same since our HD800 are after all, different pair of headphones), unless your DAC/AMP's Frequency Response deviates from RME too much.

What I did was:

1. Generate measurement result with REW;
2. Compare it with Harman Target Curve;
3. Sets the frequency and estimated magnitude of each adjusting point;
4. Generate measurement Result after EQ;
5. (Repeat step 2-4);
6. Once the output of the EQed Headphone is close to the target curve, start subjective tuning;
7. Adjust the frequency, magnitude and Q level of each adjusting point, trying to find the best preference of each adjusting point;
8. Once the EQed headphone sound as pleasing as it could get, run the measurement again;(we get preset V1 here)
9. Now use this "pleasing target curve" - which is close but not identical to the universal Harman target curve - as the new target curve;
10. Create a set of new adjusting point, this time targeting the new curve generated in step 8;
11. Trying to use minimum number of setpoints to achieve the same target curve (as we know, less is more)
12. (Now you could see why it took me so long - A target curve can be roughly achieved by infinite combination of adjustment;
13. After the new target curve is roughly reached with less number of adjusting point, do step 6-8;(we get preset V2 here)
14. Compare preset V2 to V1, if V1 sounds better, repeat step 9-13; if V2 sounds better, repeat step 9-13 but this time, use V2 as target curve instead of V1;
15. Repeat 9-14 until I couldn't make the preset sounds any better, and could not reduce the number of adjusting point anymore, we get the final preset Vn.

Note I only used Harman Target Curve in step 2 as it is not only the driver itself creating the frequency response. There's also resonance and cancellation. So if we just force the combined Frequency response to match the Harman Curve you're guaranteed to create weird sound.
The remaining works needs to be done by our own ears.

This process took me several weeks as the critical listening part could be difficult - you could only focus on analysing the sound for a limited duration.
I've been using the final preset for a while and it just sounds better than any other EQ (for HD800) I've created or found online.

I've incorporated the EQ into 2 test track which can be downloaded here:
Male Vocal with natural instruments:
Before EQ: Roadhouses & Automobiles
After EQ: Roadhouses & Automobiles - EQ
Synthesizer with symphony orchestra:
Before EQ: Main theme from Pacific Rim
After EQ: Main theme from Pacific Rim - EQ
Femel Vocal with synthesized&natural instruments:
Before EQ: Glem ikkje - Kari Bremnes
After EQ: Glem ikkje - Kari Bremnes - EQ

HD800 owners could immediately tell the difference between the 2 tracks. Please don't try the EQed tracks on other headphone/IEMs, they would only make things worse.
As Fabfilter is not opensource and actually quite expensive to purchase, I would not post the preset file here (Fabfilter use their own file format), to prevent any allegation as if I'm promoting this software. If you happen to have fabfilter and want the preset file, please PM me and I'll send privately.
If you're interested to try more tracks please upload and post the link here, and I'll incorporate the preset so you could compare with your favourite music.
Please let me know you feel about the effect of this EQ preset.

Measurement of the right channel of my HD800 without EQ (Red) and with the final preset (Green).

-qyy920

 

[LaLaLard]

P.S. The greatest difference emerges when you listen to the EQed version with maximum volume you could take, and switch back to the original tracks. This is when you would really feel the harshness of HD800 - with the original track.
P.S.2. What works for me might not for others. My taste of playback devices are natural, neutral sound. I tuned my HD800 based on my experience learning instrument (violin) and attended orchestra performance in youth. It might not be an "ideal" tuning for all music genre, let alone all other people. But I've found it works very well for me on all music genre since I arrived at this preset.

 

[solderdude]

Did you use SBAF compensation ?
The compensation supplied with the EARS is incorrect.

 

[LaLaLard]

Did you use SBAF compensation ?
The compensation supplied with the EARS is incorrect.

I used HEQ file supplied with the EARS initially but I only needed the measurement result for confirmation of any anticipated changes. The tuning was mostly done with critical listening.
I think some of the peak/trough were caused by resonance and/or other physical issue, therefore I wasnt targeting the flattest curve since the beginning as a flat curve in this case would not guarantee good sound.
Do you know where to find the SBAF compensation file? I am inclined to remeasure with that compensation file.

 

[solderdude]

You may have to register there. Maybe @maverickronin can help ?

 

[maverickronin]

I haven't played around with it but I think the latest info is here.

https://docs.google.com/spreadsheets/d/1uUqkpD3EJMzpl8FeJbNOkjApvMKAIL3IP8Y1At_ldng/edit#gid=0

It looks like they have it as a set of compensations you apply to your own EARS's calibration.

 

[Robbo99999]

What about just using an Oratory1990 Harman EQ for that headphone. It's listed here: https://www.reddit.com/r/oratory1990/wiki/index/list_of_presets

Apparently his measurement rig is one of the best you can get (or the best?), plus you're getting his measurement & EQ expertise and he is an 'audio professional' by trade. I think he's the most highly respected headphone measurer & EQ producer on the net, in my experience.

 

[LaLaLard]

What about just using an Oratory1990 Harman EQ for that headphone. It's listed here: https://www.reddit.com/r/oratory1990/wiki/index/list_of_presets

Apparently his measurement rig is one of the best you can get (or the best?), plus you're getting his measurement & EQ expertise and he is an 'audio professional' by trade. I think he's the most highly respected headphone measurer & EQ producer on the net, in my experience.

Did not know this, thanks for mentioning!
Tried his 3 presets for HD800 immediately. Impressive result, but personally I find the sound of my setup much more neutral. Oratory1990's Harman AE/OE preset for HD800 is -to some extent- good for pop and perhaps, rock, but it certainly doesn't sound that neutral. I tried his setup with some symphony aside from vocals and confirmed that.
I think he surely has one of the best test-rig in the world, but considering the massive number of headphone he tested and EQed, I doubt if he could spend much time tuning them by ears.
Having said this, I think his preset is probably a good base to start with. Lots of adjustment after critical listening might improve it.
Maybe give his filter a go with the song I uploaded and do some comparison and let me know what your thoughts are.

 

[KSTR]

I did an all-negative-gain setup, in order to prevent from clipping.

This prevents only the obvious type of clipping, but the other kind (coming from phase shift by the EQs changing the waveform) still can and will happen unless you back off another few dB's

 

[Theriverlethe]

What about just using an Oratory1990 Harman EQ for that headphone. It's listed here: https://www.reddit.com/r/oratory1990/wiki/index/list_of_presets

Apparently his measurement rig is one of the best you can get (or the best?), plus you're getting his measurement & EQ expertise and he is an 'audio professional' by trade. I think he's the most highly respected headphone measurer & EQ producer on the net, in my experience.

He is definitely a skilled acoustician, but there are limitations to his measurement and the Harman curve itself. You might still get a better subjective result with careful tuning to your individual HRTF.

 

[LaLaLard]

This prevents only the obvious type of clipping, but the other kind (coming from phase shift by the EQs changing the waveform) still can and will happen unless you back off another few dB's

That's why Fabfilter Pro-Q was my choice, instead of RME's built-in EQ or other EQ on PC&Mac. One of the three modes it has is "Linear Phase", which prevents phase from shifting.

 

[Robbo99999]

Did not know this, thanks for mentioning!
Tried his 3 presets for HD800 immediately. Impressive result, but personally I find the sound of my setup much more neutral. Oratory1990's Harman AE/OE preset for HD800 is -to some extent- good for pop and perhaps, rock, but it certainly doesn't sound that neutral. I tried his setup with some symphony aside from vocals and confirmed that.
I think he surely has one of the best test-rig in the world, but considering the massive number of headphone he tested and EQed, I doubt if he could spend much time tuning them by ears.
Having said this, I think his preset is probably a good base to start with. Lots of adjustment after critical listening might improve it.
Maybe give his filter a go with the song I uploaded and do some comparison and let me know what your thoughts are.

He is definitely a skilled acoustician, but there are limitations to his measurement and the Harman curve itself. You might still get a better subjective result with careful tuning to your individual HRTF.

I'm not sure you can get much better than the Harman Curve for headphones, it was created using research from Toole & Olive, according to my current knowledge it's the best there currently is in terms of target for headphones.

Regarding miniDSP EARS, last time I looked into it, it was acknowledged that there are major shortcomings with that bit of kit and instead Oratory1990's setup does not have the flaws of the miniDSP EARS. I'm being vague because I can't remember the details, I just remember mentally cataloguing that tuning based on miniDSP EARS was not really valid.

@Theriverlethe , do you mean tuning your headphones using white noise at different frequencies so that is of equal loudness through the frequency range. (or do I mean pink noise, I remember someone saying once you can tune to equal loudness through the frequency range, which is essentially tuning a flat curve, at least how your own ear would perceive it). I suppose the problem with that is that it doesn't take into account any hearing loss or deficiencies you have - I mean that it would boost the frequencies where you have hearing issues, which would be 'unnatural', ie say live music would not be boosted & tailored to your own hearing deficiencies, so you're not experiencing 'reality' so to speak if you tune by ear. ?

 

[Theriverlethe]

I'm not sure you can get much better than the Harman Curve for headphones, it was created using research from Toole & Olive, according to my current knowledge it's the best there currently is in terms of target for headphones.

Regarding miniDSP EARS, last time I looked into it, it was acknowledged that there are major shortcomings with that bit of kit and instead Oratory1990's setup does not have the flaws of the miniDSP EARS. I'm being vague because I can't remember the details, I just remember mentally cataloguing that tuning based on miniDSP EARS was not really valid.

@Theriverlethe , do you mean tuning your headphones using white noise at different frequencies so that is of equal loudness through the frequency range. (or do I mean pink noise, I remember someone saying once you can tune to equal loudness through the frequency range, which is essentially tuning a flat curve, at least how your own ear would perceive it). I suppose the problem with that is that it doesn't take into account any hearing loss or deficiencies you have - I mean that it would boost the frequencies where you have hearing issues, which would be 'unnatural', ie say live music would not be boosted & tailored to your own hearing deficiencies, so you're not experiencing 'reality' so to speak if you tune by ear. ?

Harman Curve is the best target curve for headphone manufacturers, but it’s just an average of listener preferences. Your own preference and/or physiology may deviate significantly. People talk about headphones removing room interaction, but you’re just creating another cavity with its own resonances and cancellations.

The problem with EQ’ing to flat-sounding noise is that it doesn’t take equal loudness contour into account. Results will just be wrong. Unless you can have your own HRTF measured, tuning to subjective preference might be as good as it gets. Harman Curve provides a good starting point.

 

[LaLaLard]

I'm not sure you can get much better than the Harman Curve for headphones, it was created using research from Toole & Olive, according to my current knowledge it's the best there currently is in terms of target for headphones.

I did not say it could get better than the Harman Curve. But it is definetely possible to get better than the EQ preset provided by Oratory1990, for a number of reason:

1. He has limited the number of the adjustment that could be applied to 10, which I found to be insufficient to create a good EQ;
2. He has targeted Harman Curve with all resonance factor ignored, in other words, it is the frequency response of the headphone plus the effect of any resonance that is targeting the Harman Curve.

Like I said, I did not doubt about Oratory1990's expertise in acoustic nor the quality of his testing gear. I just want to say it is still possible to generate a different EQ setting that sounds better than just copy and paste his setup.

 

[LaLaLard]

Harman Curve is the best target curve for headphone manufacturers, but it’s just an average of listener preferences. Your own preference and/or physiology may deviate significantly. People talk about headphones removing room interaction, but you’re just creating another cavity with its own resonances and cancellations.

The problem with EQ’ing to flat-sounding noise is that it doesn’t take equal loudness contour into account. Results will just be wrong. Unless you can have your own HRTF measured, tuning to subjective preference might be as good as it gets. Harman Curve provides a good starting point.

This is exactly the point I was trying to make. For what I need - which is getting instant, objective feedback on my adjustment - MiniDSP EARS is sufficient. I did not want my HD800 and its resonance + cancellation to land 100% on the universal Harman Curve, nor did I think doing that would create the best sound. A testing Rig could only take us so far, the remaining road needs to be walked with our own ears.

 

[Robbo99999]

Harman Curve is the best target curve for headphone manufacturers, but it’s just an average of listener preferences. Your own preference and/or physiology may deviate significantly. People talk about headphones removing room interaction, but you’re just creating another cavity with its own resonances and cancellations.

The problem with EQ’ing to flat-sounding noise is that it doesn’t take equal loudness contour into account. Results will just be wrong. Unless you can have your own HRTF measured, tuning to subjective preference might be as good as it gets. Harman Curve provides a good starting point.

Harman Curve is based on average preference, but I believe it's based on average preference around an initial Diffuse Field curve (ie one that was created to simulate flat speakers in a room taking an average HRTF into account), so I think bass level was tuned on top of the Diffuse Field curve. So I would think Harman Curve is a good start then you tweak bass to your liking (using broad Low Shelf), and perhaps tune 10kHz+ with broad High Shelf filter too. It sounds like we agree on that broadly.

Yes, I agree about tuning to flat sound and the problems there, that is what I was getting at.

 

[Robbo99999]

I did not say it could get better than the Harman Curve. But it is definetely possible to get better than the EQ preset provided by Oratory1990, for a number of reason:

1. He has limited the number of the adjustment that could be applied to 10, which I found to be insufficient to create a good EQ;
2. He has targeted Harman Curve with all resonance factor ignored, in other words, it is the frequency response of the headphone plus the effect of any resonance that is targeting the Harman Curve.

Like I said, I did not doubt about Oratory1990's expertise in acoustic nor the quality of his testing gear. I just want to say it is still possible to generate a different EQ setting that sounds better than just copy and paste his setup.

I'm not sure it's possible to effectively EQ a headphone beyond using a Harman Curve that has been done on a good rig (like Oratory's)....and then you might tweak the bass and treble with broad Low Shelf filters to taste, so sub 100Hz for the Low Shelf, and then around 10kHz+ for the High Shelf. He says in his pdf's that you can tune those filters to taste e.g. in this one he says as much (https://www.dropbox.com/s/dm0m6u3s3b4zqzl/Sennheiser HD600.pdf?dl=0 ). I don't think you can get much better than that because otherwise I think a person is just "wading around in the dark". Olive & Toole agree that bass and treble should be adjusted to taste using broad tone controls to account for different recordings & personal preference, but they believe in keeping the "core" the same.

 

[JIW]

Did not know this, thanks for mentioning!
Tried his 3 presets for HD800 immediately. Impressive result, but personally I find the sound of my setup much more neutral. Oratory1990's Harman AE/OE preset for HD800 is -to some extent- good for pop and perhaps, rock, but it certainly doesn't sound that neutral. I tried his setup with some symphony aside from vocals and confirmed that.
I think he surely has one of the best test-rig in the world, but considering the massive number of headphone he tested and EQed, I doubt if he could spend much time tuning them by ears.
Maybe give his filter a go with the song I uploaded and do some comparison and let me know what your thoughts are.

In what way was his preset less neutral than yours?

The gain for the bass shelf (below 120 Hz or so) (filter 2) is a matter of preference. For more linear extension, try reducing the gain of filter 2 by 3 dB. Similarly, since the measurements are accurate only up to 8 kHz or so, the filtering in the top octave is also a matter of preference. However, rather than using notch/bell filters that are highly dependent on the geometry between the headphone and ear/recording device which can change significantly with positioning of the headphone, I suggest you try a high shelf (maybe a notch centered at the top of the hearing range to get desired roll-off may also do). How you achieve the shelf matters much less than that it does not emphasize narrow frequency ranges.

Also, I see that your preset has more gain between 2 and 3 kHz. As indicated by oratory1990, try adding some gain (maybe 2-3 dB) to filter 6. Further, I see that you did not cut the broad hump between 100 and 900 Hz while at the same not increasing the sub-bass. Have you already tried that?

Overall, I think your preset is actually quite close to the Harman target given personal preference and the limitations of the measurement equipment. However, curiously, the infamous 6 kHz resonance is not much reduced in your preset only what surrounds is - arguably - attenuated excessively. This may be entirely due to differences in measurement setup and oratory1990's is much more accurate in that region.

 

[Robbo99999]

As a quick side note, does anybody know which country Oratory1990 lives? I'm thinking of sending him my AKG K702 headphones, but I had a feeling he lived in the USA? (Hopefully he's in Europe).

 

[LaLaLard]

In what way was his preset less neutral than yours?

The gain for the bass shelf (below 120 Hz or so) (filter 2) is a matter of preference. For more linear extension, try reducing the gain of filter 2 by 3 dB. Similarly, since the measurements are accurate only up to 8 kHz or so, the filtering in the top octave is also a matter of preference. However, rather than using notch/bell filters that are highly dependent on the geometry between the headphone and ear/recording device which can change significantly with positioning of the headphone, I suggest you try a high shelf (maybe a notch centered at the top of the hearing range to get desired roll-off may also do). How you achieve the shelf matters much less than that it does not emphasize narrow frequency ranges.

Also, I see that your preset has more gain between 2 and 3 kHz. As indicated by oratory1990, try adding some gain (maybe 2-3 dB) to filter 6. Further, I see that you did not cut the broad hump between 100 and 900 Hz while at the same not increasing the sub-bass. Have you already tried that?

Overall, I think your preset is actually quite close to the Harman target given personal preference and the limitations of the measurement equipment. However, curiously, the infamous 6 kHz resonance is not much reduced in your preset only what surrounds is - arguably - attenuated excessively. This may be entirely due to differences in measurement setup and oratory1990's is much more accurate in that region.

You could listen to the songs I posted and try with Oratory’s settings yourself if you have a pair of HD800. Unless the original recording is awkward, I know which one sounds more natural.
Also my setup is indeed created with negative gain only. The reason that the curves crosse each other was simply that I ran the measurements at slightly different level aiming to indicate how sound signature really changes - when you are using pre-gain to match the overall level on top of the EQ. No one would listen to an all negative filter and think it sounds better than the original one as level difference actually lead one to think the sound is “better” or “fuller”, which all folks here know is not correct, right?