• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Which preset sounds best?

  • 1: ASR

    Votes: 3 12.0%
  • 2: Crinacle

    Votes: 3 12.0%
  • 3: oratory1990 Grapher

    Votes: 4 16.0%
  • 4: oratory1990 PDF

    Votes: 16 64.0%
  • 5: Resolve

    Votes: 0 0.0%
  • 6: rtings

    Votes: 1 4.0%
  • 7: SSN

    Votes: 0 0.0%

  • Total voters
    25

staticV3

Master Contributor
Joined
Aug 29, 2019
Messages
7,529
Likes
11,995
Here's a little experiment to see which HE400se measurement produces the best EQ preset when corrected to Harman OE2018.

Note that this headphone has impressively low unit variation, which might make deciding rather difficult, so I'm giving you three votes.
HiFiMan HE400se Frequency Response Comparison.png

The presets can only be used with vanilla Equalizer Apo for Windows. If you're using Peace, then you can disable that temporarily.
Here's how to load the presets and A/B them: YouTube Link
Screenshot 2022-10-27 014644.png

Note that in the video you see me setting the preamp to -8.5dB. Here, -10.7dB are required!
rtings' is the only non-GRAS measurement of the bunch, so I "converted" it to GRAS using Jaakko's case study. However, even after compensation it asked for 20+dB boost above 10kHz, so I cut its preset short at about 9.5k
Please make sure that no other DSP is active while comparing.
In case you're interested, here's why I'm using variable band GEQ instead of PEQ. It produces nearly identical results regardless of sample rate. PEQ doesn't.
If you want to compare EQ On vs EQ Off, then you can toggle just the EQ without the overall volume changing - another advantage over Peace.

Happy listening!
 

Attachments

  • HE400se GEQ.zip
    26.4 KB · Views: 915

GaryH

Major Contributor
Joined
May 12, 2021
Messages
1,348
Likes
1,804
In case you're interested, here's why I'm using variable band GEQ instead of PEQ. It produces nearly identical results regardless of sample rate. PEQ doesn't.

Filter warping by sample rate is marginal, especially at the lower sample rates:

En2pCUd.png

trpokjJ.png

gVM6rOq.png

I'd use PEQ, and just tell people to make sure their sample rate is set at 44.1 or 48 kHz. I'd be more concerned about potential audible ringing when using a 600-band GEQ...
 

GaryH

Major Contributor
Joined
May 12, 2021
Messages
1,348
Likes
1,804
Oratory's results differ. Are you sure your methodology is sound? And can you post the individual filters used for the above preset? Anyway as I said, just tell people to make sure their sample rate is set to the standard 44.1 or 48 kHz, problem solved (without introducing possible audible issues related to using GEQ).
 
Last edited:

GaryH

Major Contributor
Joined
May 12, 2021
Messages
1,348
Likes
1,804
Well we won't know exactly unless you post the individual filters for the above preset so we can can see whether they're actually a reasonable example. And even better redo it with the exact same single filters as Oratory's examples above to compare like for like.
 
OP
staticV3

staticV3

Master Contributor
Joined
Aug 29, 2019
Messages
7,529
Likes
11,995
unless you post the individual filters for the above preset
I'm not following tbh. What filters? The presets are not based on any parametric filters. It's literally measured FR (+some HF smoothing) minus target FR. Here's oratory's Grapher FR for example:

1. Raw Frequency Response
oratory grapher raw.png
2. HF Smoothing
oratory grapher smoothed.png
3. overlay Harman
oratory grapher smoothed vs Harman.png
4. calculate Error
oratory grapher smoothed error.png
5. invert Error
oratory grapher smoothed GEQ.png

And that green graph is literally the graphic equalizer you're applying in EQApo.
 
OP
staticV3

staticV3

Master Contributor
Joined
Aug 29, 2019
Messages
7,529
Likes
11,995
I'm talking about this:
Ah, my bad. Here are the filters:
Filter 1: ON PK Fc 77 Hz Gain 4.0 dB Q 0.15
Filter 2: ON PK Fc 163 Hz Gain -9.6 dB Q 0.70
Filter 3: ON PK Fc 1337 Hz Gain 4.1 dB Q 3.21
Filter 4: ON PK Fc 2513 Hz Gain 7.3 dB Q 1.88
Filter 5: ON PK Fc 3529 Hz Gain -2.4 dB Q 1.14
Filter 6: ON PK Fc 10930 Hz Gain 2.6 dB Q 1.79
Filter 7: ON PK Fc 19886 Hz Gain -7.7 dB Q 0.46

Just a random PEQ I did for someone. And yes, I'm quite sure my methodology is sound.
 
OP
staticV3

staticV3

Master Contributor
Joined
Aug 29, 2019
Messages
7,529
Likes
11,995
Results using the exact same single filters as Oratory above would confirm (or deny) this.
be my guest. You have everything you need to prove my methodology not sound.
 

GaryH

Major Contributor
Joined
May 12, 2021
Messages
1,348
Likes
1,804
It's your methodology to confirm or deny by reproducing the results from a professional acoustic engineer, not mine.
 
OP
staticV3

staticV3

Master Contributor
Joined
Aug 29, 2019
Messages
7,529
Likes
11,995
@GaryH
You question my methodology, you claim that PEQ FR depencence on Fs is negligible, you claim that GEQ may produce audible ringing.
I think it's time you put your money where your mouth is back those claims up.
 

GaryH

Major Contributor
Joined
May 12, 2021
Messages
1,348
Likes
1,804
The fact that you're unwilling to reproduce professional results to confirm yours is concerning. The onus is on you to backup your claim that your methodology is sound. 'Assume I'm correct until someone else proves me wrong' is not scientific. The scientific way is to first prove and demonstrate your methodology is sound before accepting its results, by making sure it corroborates existing correct results from professionals in the field. There's no way to be 100% sure the conditions, software and hardware I use would be exactly the same as yours so no, I can't do this for you.
 
Last edited:
OP
staticV3

staticV3

Master Contributor
Joined
Aug 29, 2019
Messages
7,529
Likes
11,995
The fact that you're unwilling to reproduce professional results to confirm yours is concerning.
The fact that you're all talk, no substance sure makes me think that I'm wasting my time on someone agruing for argumen't sake.
The onus is on you to backup your claim that your methodology is sound.
Here's my methodology: YouTube Link
Measuring the analog output of the DAC as well as using another PEQ program produced identical results btw:
Mathaudio_Filter_response_vs_Sample_rate.png
but staying digital is more convenient.
There's no way to be 100% sure the conditions, software and hardware I use would be exactly the same as yours so no, I can't do this for you.
So how would I make 100% sure the conditions, software and hardware I use would be exactly the same as what oratory used?
 

GaryH

Major Contributor
Joined
May 12, 2021
Messages
1,348
Likes
1,804
So how would I make 100% sure the conditions, software and hardware I use would be exactly the same as what oratory used?
You don't necessarily need to, if your results can replicate his exactly. This should be the first thing you do before presenting any other results.
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
I used a treble peak filter and added some low bass. Liked it that way but it is still entry level planar.
So my vote would be none of the above ;-)
 
OP
staticV3

staticV3

Master Contributor
Joined
Aug 29, 2019
Messages
7,529
Likes
11,995
You don't necessarily need to, if your results can replicate his exactly. This should be the first thing you do before presenting any other results.
you're actually trolling now, aren't you? What as waste of time.
 
OP
staticV3

staticV3

Master Contributor
Joined
Aug 29, 2019
Messages
7,529
Likes
11,995
I'm not trolling at all. Reproducing prior professional results to confirm a methodology is basic science. The fact that you're spending time with this back and forth but won't do this speaks volumes.
I've shown you the measurements using digital caputure, a four minute video showing exactly how I did that capture, plus a graph showing what the response looks like if instead I capture the analog output using another PEQ program entirely. All of those match each other.
Meanwhile, what have you contributed to this discussion? Any valuable data? Anything at all?

I think it speaks volumes that you're willing to dismiss all of that just because it doesn't match exactly what your favorite internet person once posted. Talk about being scientific :rolleyes:
 
Last edited:

Music1969

Major Contributor
Joined
Feb 19, 2018
Messages
4,636
Likes
2,808
@staticV3 maybe you can try my EQ.

Basically I applied oratory1990's 'linear bass extension' EQ (see the notes in his Sundara PDF) to Crinacle.

To me, this EQ best fits all Sundara measurements I've seen and leaves out the outlier 'issues'

Preamp: -9.0 dB Filter 1: ON LSQ Fc 28 Hz Gain 6.00 dB Q 0.71 Filter 2: ON PK Fc 66 Hz Gain -0.80 dB Q 1.50 Filter 3: ON LSQ Fc 105 Hz Gain 3.50 dB Q 0.71 Filter 4: ON PK Fc 940 Hz Gain -1.0 dB Q 3.70 Filter 5: ON PK Fc 2250 Hz Gain 3.30 dB Q 1.40 Filter 6: ON PK Fc 6105 Hz Gain 2.2 dB Q 1.50 Filter 7: ON HSQ Fc 10000 Hz Gain -1.0 dB Q 0.71
 
Top Bottom