• 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!

Help equalizing the Shure SRH1540

q3cpma

Major Contributor
Joined
May 22, 2019
Messages
3,060
Likes
4,416
Location
France
Hello,

After the cable on my DT250 snapped, I used it as an excuse to buy this model that seems perfect with "a bit" of EQ. But not all sources seem to agree on the measurement side, and I plan on using maximum 3 bands (ideally 2) to save a bit of battery, since this is a Rockbox setup.
Here's what the AutoEQ repository shows:

Crinnacle (GRAS 43AG-7):
Shure%20SRH1540.png


headphone.com:
Shure%20SRH1540.png


InnerFidelity:
Shure%20SRH1540.png


Oratory1990:
Shure%20SRH1540.png


@solderdude:
index.php


Innerfidelity's PDF:
index.php


So, who to believe? All show a massive LF bump that'll go away with a low Q filter, but Crinnacle and Oratory have a different result under 60 Hz, only Oratory gets this 2.5~3 kHz peak, while @solderdude shows a marked lack of HF (and talk about it in his subjective evaluation). Confusingly, Innerfidelity's own PDF also seem to use a strange target curve, as the raw difference of level between 1 kHz and that peak is ~12 dB, which seem to be enough to me (and to both Harman and the author of AutoEQ).

Any opinion on this?
 

Attachments

  • diyaudioheaven.png
    diyaudioheaven.png
    40.5 KB · Views: 2,548
  • innerfidelity.png
    innerfidelity.png
    72.4 KB · Views: 2,642

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
Innerfidelity plots have always been incorrect as he used an incorrect compensation.
This 'target' will get you closer to reality (it is not an exact target)
comp-curve-if1.png

I don't use HATS because I am particularly interested in 6kHz and higher.

Just try all EQ and see which one sounds most believable to you.
Then work out a simple EQ that gets you there.
 
Last edited:
OP
q3cpma

q3cpma

Major Contributor
Joined
May 22, 2019
Messages
3,060
Likes
4,416
Location
France
Going to use newer (03/2021) Oratory1990, in fine, at least for the LF. I'll experiment a little for the 2.5~3 kHz area, but that's all.
 

staticV3

Master Contributor
Joined
Aug 29, 2019
Messages
7,536
Likes
12,004
So, who to believe?
Your own ears.
In my experience, no one source of EQ settings can be trusted unanimously.
For one headphone rtings might sound best and for another, rtings' preset may sound awful with oratory taking the cake.
My recommendation is always to take every preset that you can find and compare them directly using something like EQApo + Peace GUI.
If you'd like, then I can help you set this up and prepare the presets for you. I can convert them all to 3-band Rockbox presets using the python script as well.
 
OP
q3cpma

q3cpma

Major Contributor
Joined
May 22, 2019
Messages
3,060
Likes
4,416
Location
France
Your own ears.
In my experience, no one source of EQ settings can be trusted unanimously.
For one headphone rtings might sound best and for another, rtings' preset may sound awful with oratory taking the cake.
My recommendation is always to take every preset that you can find and compare them directly using something like EQApo + Peace GUI.
If you'd like, then I can help you set this up and prepare the presets for you. I can convert them all to 3-band Rockbox presets using the python script as well.
You're right that I should try a bit on the computer before settling down.
> EQApo + Peace GUI
On Linux but not using Pulse nor Pipewire (though this one seems promising), so I'll use ffmpeg with the anequalizer/lowshelf/highshelf filters.

Thanks for your proposed help, but I'll manage.
 
Last edited:

staticV3

Master Contributor
Joined
Aug 29, 2019
Messages
7,536
Likes
12,004
@q3cpma wouldn't it be possible to upload all presets as Rockbox EQ files to your MP3 player and A/B them in there?
FWIF, I've only recently started playing around with the AutoEQ python script and totally wouldn't mind converting a bunch of presets into 10/3/2-band PEQ Rockbox .cfg files, just to get used to the workflow.
And if there's someone benefitting from my experimentation, even better, right? :D
 
OP
q3cpma

q3cpma

Major Contributor
Joined
May 22, 2019
Messages
3,060
Likes
4,416
Location
France
It would be, but why? It'll probably be faster to switch on the computer; in fact, I'll use mpv to be able to switch the EQ parameters during the track.
 

staticV3

Master Contributor
Joined
Aug 29, 2019
Messages
7,536
Likes
12,004
It would be, but why? It'll probably be faster to switch on the computer; in fact, I'll use mpv to be able to switch the EQ parameters during the track.
I assumed that MPV/ffmpeg wouldn't be able to import AutoEQ's .txt files, which would then necessitate punching in all parameters manually, which, with the seven available measurements that I could find, would be quite bothersome.
That in turn would make solutions like Peace or Rockbox with direct file imports preferable.
 
Last edited:

Χ Ξ Σ

Senior Member
Forum Donor
Joined
Jul 16, 2020
Messages
457
Likes
1,969
Location
UTC-8
Better than pink noise? Probably easy to do both with SoX.
This method allows you to identify resonance peaks while the headphones are on you head. As you can imagine your ears and the headphone measurement gears are not the exactly same. Measurement gears can record resonances that are not there when you actually listen, and it often goes the other way, too. All the EQ profiles are a good start, but listening to the sweep helps you fine tune it and allows you to achieve the smoothest response in your ears.
 
OP
q3cpma

q3cpma

Major Contributor
Joined
May 22, 2019
Messages
3,060
Likes
4,416
Location
France
I assumed that MPV/ffmpeg wouldn't be able to import AutoEQ's .txt files, which would then necessitate punching in all parameters manually, which, with the seven available measurements that I could find, would be quite bothersome.
That in turn would make solutions like Peace or Rockbox with direct file imports preferable.
Let's say that I'm not busy to the point I can't make a quick sh script.
 

daftcombo

Major Contributor
Forum Donor
Joined
Feb 5, 2019
Messages
3,687
Likes
4,068
IMO, the 8Hz dip is too narrow to be EQed. And the surrounding peaks too.
I would simply put a high-pass filter up to 200 Hz to begin with.

(I was watching Solderdude's measurements.)
 

Feelas

Senior Member
Joined
Nov 20, 2020
Messages
390
Likes
315
It would be, but why? It'll probably be faster to switch on the computer; in fact, I'll use mpv to be able to switch the EQ parameters during the track.
Remember to let any eventual changes settle for a week or two without switching around, since then you'll be very hard pressed to decide what is better. Two weeks seems to be fine for proper accomodation to a new FR.
 

staticV3

Master Contributor
Joined
Aug 29, 2019
Messages
7,536
Likes
12,004
Remember to let any eventual changes settle for a week or two without switching around, since then you'll be very hard pressed to decide what is better. Two weeks seems to be fine for proper accomodation to a new FR.
That's pretty counterproductive when it comes to comparing different frequency response, don't you think?
 

Feelas

Senior Member
Joined
Nov 20, 2020
Messages
390
Likes
315
That's pretty counterproductive when it comes to comparing different frequency response, don't you think?
It seems counterproductive, but read up about how producers end up in a circle of tweaking stuff and then at the second day end up being very disillusioned by the changes they made. You can of course disregard the aspect of letting yourself accomodate, but from personal experience I'd stray from saying that it is counterproductive. Playing a lot with EQ, on the other hand, is especially annoying when tweaking something new and clicking around all the time. Pretty much you're gonna end up liking one EQ one day and a different one on the second day. If letting yourself stay for a week or two, the reults might be entirely different.

Head has to make the adjustment towards new FR and it takes time. Make out what you want from that. Doing EQ corrections on speakers or headphones is like getting a new pair of ears. I could dig out an interesting study where it was measured how long it takes people to accomodate to fake ears (w/ new earshape) in terms of perceiving the sound direction, if you're interested.
 

daftcombo

Major Contributor
Forum Donor
Joined
Feb 5, 2019
Messages
3,687
Likes
4,068
It seems counterproductive, but read up about how producers end up in a circle of tweaking stuff and then at the second day end up being very disillusioned by the changes they made. You can of course disregard the aspect of letting yourself accomodate, but from personal experience I'd stray from saying that it is counterproductive. Playing a lot with EQ, on the other hand, is especially annoying when tweaking something new and clicking around all the time.

Head has to make the adjustment towards new FR and it takes time. Make out what you want from that. Doing EQ corrections on speakers or headphones is like getting a new pair of ears. I could dig out an interesting study where it was measured how long it takes people to accomodate to fake ears (w/ new earshape) in terms of perceiving the sound direction, if you're interested.
Producers don't know how it should sound when creating new tracks. Here, one can EQ until it sounds as expected compared to previous pairs of headphones or speakers.
 
Top Bottom