[KinGensai]

Hi, I'm curious about how everyone utilizes EQ. Specifically, how to achieve changes to the sound image to correct percievable errors.

I recently figured out that, for me, female vocals and guitars adjust their height in the image by adjusting 1.5k and 2.9k. This plays out consistently across my IEMs and headphones, so I am now wondering if there are other things like this that people know about.

If there is some kind of resource for this sort of knowledge, I'd appreciate that too.

 

[Ufanco]

I know this answer will get flack but here goes, I don’t use EQ in my system instead I buy an iem that closely matches frequency I enjoy. Then match rest of components to give the last tweaks to sound. I use a Ares ii and tube amp to find tune the final sound.

First would be a find a good EQ app then try find a good online guild. Only one I sometime use is neutron player but sure others will add there favorites.

Here is a graph that might help if use EQ

 

[megabigeye]

Here's a guide for EQing.

 

[bigshot]

I learned to EQ by just experimenting. Get a good graphic equalizer with a lot of bands and just play with it as you listen to music. Figure out what the frequency bands relate to in actual sound, and figure out what part of the music lives in those bands. It's a process of just learning the names for sound. After a while, you'll hear an instrument and just know the frequency range it falls within. Then you can listen to a wide range of music and start analyzing what seems too prominent and what seems to be too recessed. Once you have your footing, you'll find that a parametric equalizer is more flexible than a graphic one. But parametric are less intuitive- harder to learn with.

My best tips for EQ are...

Get as close to your ideal with your transducers before equalizing. EQ is best to do fine tuning, not gross adjustments for wildly imbalanced transducers.

Equalize subtractively to avoid clipping. Don't boost. If something seems too recessed, pull everything around it back. It's not intuitive, but sometimes you'll find that a frequency band that is recessed isn't recessed at all- it may be just masked by a peak an octave above. Cutting back an octave higher might effectively boost an octave before. Make note of where the octave breaks are so you can do this trick. It's not strictly double the number.

When you're searching for your own personal target curve, start from a fixed point, like Harman or measured flat with speakers. Then make small incremental changes and live with them and hear how they affect lots of music before making another change. If it doesn't work, go back to your fixed point and start again. The slower you do it, the less likely you'll randomly EQ yourself into a weird corner.

 

[castleofargh]

Hi, I'm curious about how everyone utilizes EQ. Specifically, how to achieve changes to the sound image to correct percievable errors.

I recently figured out that, for me, female vocals and guitars adjust their height in the image by adjusting 1.5k and 2.9k. This plays out consistently across my IEMs and headphones, so I am now wondering if there are other things like this that people know about.

If there is some kind of resource for this sort of knowledge, I'd appreciate that too.

Perceived elevation is based on your own head and ears and how sound coming from a certain angle is "EQed" by your body. So there are going to be some generalities common to humans with non-triangular shaped head and who didn't spend a decade getting their outer ears crushed with wrestling or other gentle activities. But it's still mostly you, your head and the settings you want/need.
The biggest issues being that the horizontal angle for the incoming sound is also causing FR variations on top of interaural timing variations. What might feel like eye level frequency response right in front of you, probably won't as the instrument is panned to one side (or vice versa).

If you're interested in working that out for frontal (mono) sounds, David Griesinger has been a strong advocate for many years of a method using equal loudness perception and a speaker in front of you as reference. It's predominantly aimed at binaural recordings, but it might still be an interesting little experiment and give you more data to fine tune your overall preferred frequency response. I tried it maybe 5 years ago when there was only a work in progress PC software with clear issues that he very kindly had shared with me after a conversation by email. Now there seems to be a more modern version, some VST, along with cellphone apps (probably the most convenient for people in general). Here is him explaining the app:

 

[KinGensai]

Thanks for the responses! This is a lot, so I'll take some time to go through it. The knowledge is much appreciated.

 

[Takeda`1537]

Hi, I'm curious about how everyone utilizes EQ. Specifically, how to achieve changes to the sound image to correct percievable errors.

I recently figured out that, for me, female vocals and guitars adjust their height in the image by adjusting 1.5k and 2.9k. This plays out consistently across my IEMs and headphones, so I am now wondering if there are other things like this that people know about.

If there is some kind of resource for this sort of knowledge, I'd appreciate that too.

Aside from the EQ, I would suggest to consider also trying with a sonic exciter / sonic maximizer. These type of processors improve sound by working with harmonics and fixing envelope distortion.

In case you are using a computer source, there is a plug in that works very well with APO Equalizer:
https://sourceforge.net/projects/equalizerapo/
https://www.audiopluginsforfree.com/la-petite-excite/

In case you have a analog audio chain (to be placed in between the source/pre-amp and the amp but always after the EQ and last in the chain), you could try with:
https://web.archive.org/web/2022012...sound.com/products/sonic-maximizers/282i.aspx

 

[KinGensai]

Perceived elevation is based on your own head and ears and how sound coming from a certain angle is "EQed" by your body. So there are going to be some generalities common to humans with non-triangular shaped head and who didn't spend a decade getting their outer ears crushed with wrestling or other gentle activities. But it's still mostly you, your head and the settings you want/need.
The biggest issues being that the horizontal angle for the incoming sound is also causing FR variations on top of interaural timing variations. What might feel like eye level frequency response right in front of you, probably won't as the instrument is panned to one side (or vice versa).

If you're interested in working that out for frontal (mono) sounds, David Griesinger has been a strong advocate for many years of a method using equal loudness perception and a speaker in front of you as reference. It's predominantly aimed at binaural recordings, but it might still be an interesting little experiment and give you more data to fine tune your overall preferred frequency response. I tried it maybe 5 years ago when there was only a work in progress PC software with clear issues that he very kindly had shared with me after a conversation by email. Now there seems to be a more modern version, some VST, along with cellphone apps (probably the most convenient for people in general). Here is him explaining the app:

Thank you, this is exactly the kind of research I was looking for. I can't find a working android app, I'll have to use the desktop version.

I have been slowly learning how to EQ as I've indulged in this hobby, and it's only been recently with squig.link that I've been able to start really understanding how FR adjustments change the less concrete parts of percieved sound like imaging.

I was primarily interested in any research pertaining to HRTF and if there are any established causal relationships between head/pinna dimensions and FR changes, but thinking about it now conjures images of a logistical nightmare lol. This app is a lot simpler, Dr. Griesinger's work is very interesting. It's a dense read, but I'll be diving in more.

 

[gregorio]

I would suggest to consider also trying with a sonic exciter / sonic maximizer. These type of processors improve sound by working with harmonics and fixing envelope distortion.

Personally I would not advise that. Sonic exciters have been around for nearly 50 years and have been widely employed in music production and mastering, in the more rock/pop genres almost ubiquitously. For this reason I would not advise adding more, the correct amount has already been added.

G

 

[KinGensai]

I learned to EQ by just experimenting. Get a good graphic equalizer with a lot of bands and just play with it as you listen to music. Figure out what the frequency bands relate to in actual sound, and figure out what part of the music lives in those bands. It's a process of just learning the names for sound. After a while, you'll hear an instrument and just know the frequency range it falls within. Then you can listen to a wide range of music and start analyzing what seems too prominent and what seems to be too recessed. Once you have your footing, you'll find that a parametric equalizer is more flexible than a graphic one. But parametric are less intuitive- harder to learn with.

My best tips for EQ are...

Get as close to your ideal with your transducers before equalizing. EQ is best to do fine tuning, not gross adjustments for wildly imbalanced transducers.

Equalize subtractively to avoid clipping. Don't boost. If something seems too recessed, pull everything around it back. It's not intuitive, but sometimes you'll find that a frequency band that is recessed isn't recessed at all- it may be just masked by a peak an octave above. Cutting back an octave higher might effectively boost an octave before. Make note of where the octave breaks are so you can do this trick. It's not strictly double the number.

When you're searching for your own personal target curve, start from a fixed point, like Harman or measured flat with speakers. Then make small incremental changes and live with them and hear how they affect lots of music before making another change. If it doesn't work, go back to your fixed point and start again. The slower you do it, the less likely you'll randomly EQ yourself into a weird corner.

Funny enough, I didn't know graphic equalizers were a thing, I started with a parametric equalizer and bumbled my way to now.

It took a lot of trial and error for me to learn everything you mentioned, and all your tips are spot on. Masking in particular was an important one that's not intuitive, but I felt it with mid bass muddying up the mids and connected the dots after reading on the topic.

I'm pretty much at my endgame for portable audio now (QDC V14) since they produce a sound that is quite near exactly what I'm looking for, so now it's up to me to fine-tune it, which is how I found the 1.5k/2.9k phenomenon.

I'm more of a humanities type of dumb, so hard sciences tend to beat me around too much lol. I appreciate you folks offering help.

 

[castleofargh]

Thank you, this is exactly the kind of research I was looking for. I can't find a working android app, I'll have to use the desktop version.

I have been slowly learning how to EQ as I've indulged in this hobby, and it's only been recently with squig.link that I've been able to start really understanding how FR adjustments change the less concrete parts of percieved sound like imaging.

I was primarily interested in any research pertaining to HRTF and if there are any established causal relationships between head/pinna dimensions and FR changes, but thinking about it now conjures images of a logistical nightmare lol. This app is a lot simpler, Dr. Griesinger's work is very interesting. It's a dense read, but I'll be diving in more.

To be honest, how do I say this... some of his views aren't to the liking of several acoustic engineers.
Nobody's perfect. But he's a cool guy interested in helping others enjoy music. Just that puts him at the top of the world in my views.
Anyway, I don't know of an approach to your question that was easier and more direct than his method with one speaker in front of you looking straight at it to fool around with perceived elevation.

In case you ever decide to go crazy about headphones and sound localization improvements, maybe have a look at another of my recent posts to someone asking about neutral headphone/IEM https://www.head-fi.org/threads/neutral-sound-practical-and-philosophical-q.967769/post-17508449 Check the last big chapter of my way too long answer for some options going from very good speaker simulation to... rolling tubes?

 

[Takeda`1537]

Personally I would not advise that. Sonic exciters have been around for nearly 50 years and have been widely employed in music production and mastering, in the more rock/pop genres almost ubiquitously. For this reason I would not advise adding more, the correct amount has already been added.

G

Guess it is about personal preference... I have been experimenting with multiple ones and all those who pass by my place and try my gears end up getting one.. Bellari, Rolls or BBE... While I would say that they have to be used with moderation and are not a replacement to EQing, I would personally not take out my exciter from my sound chain. Music just sounds better, ... but I say it is a matter of personal listening preferences.

 

[gregorio]

I would personally not take out my exciter from my sound chain.

That’s entirely your prerogative.

Music just sounds better, ...

No, generally music would sound worse. If it did actually sound better then the mix or mastering engineers would have added it (or added more of it) and/or the producer would have requested more. An obvious place where it would likely sound better would be old, unrestored analogue recordings, because HF response deteriorates over time/usage with analogue media. I say “unrestored” because the use of exciters is common/typical when restoring old analogue recordings, so again, any additional application would not be advisable.

but I say it is a matter of personal listening preferences.

Sure, if you have the personal preference of adding distortion/changing the content, that’s entirely up to you but then it’s not just a personal listening preference if you’re publicly advising it’s use to others.

G

 

[Takeda`1537]

I am still far from convinced.. Understand your logic but evidence, to my hears sounds better. Given that there is no holy sound type, sound editing (or its avoidance) is still a personal preference. I like it with, you prefer it without and both positions are valid and stand for for one and are wrong for the other.
If it sounds good to me, if I have empirical evidence of friends getting one too once they hear it how music sounds on my kit (not the full sample but a good majority) on what type of basis are you saying that I may not share my experience with others?
Your listening preference is worth as much as mines and each of us is keeping his/her own...
F

 

[KinGensai]

I am still far from convinced.. Understand your logic but evidence, to my hears sounds better. Given that there is no holy sound type, sound editing (or its avoidance) is still a personal preference. I like it with, you prefer it without and both positions are valid and stand for for one and are wrong for the other.
If it sounds good to me, if I have empirical evidence of friends getting one too once they hear it how music sounds on my kit (not the full sample but a good majority) on what type of basis are you saying that I may not share my experience with others?
Your listening preference is worth as much as mines and each of us is keeping his/her own...
F

Well, when I tried the links you provided, my audio started peaking and distorting the upper mids, especially on my metal library, so nah. I'm after precision and fidelity to the recording. Thanks for the suggestion though.