How does one go about digitally tuning a pair of open backs or IEM's or closed cans?
May 28, 2020 at 5:36 PM Thread Starter Post #1 of 3

punkedrock

100+ Head-Fier
Joined
Apr 2, 2018
Posts
150
Likes
30
Location
South East, USA
Would one just want to achieve the flattest response from somebody else's chart? Or do people actually manage to measure their own headphones on the cheap?

I found the following website: https://www.rtings.com/headphones/1-4/graph#816/3992
Now based of this graph for the Sennheiser HD660S would I just adjust my DAC/AMP's EQ settings (FiiO Q5S) or my Sony Walkman NW-ZX507 (Arriving tomorrow.) and try and have the flattest response possible?

Well in car audio we usually want a downward curve on the graph, so I'd like some input if I should be going for this or a flat curve... I'm very inexperienced in tuning in general and even more so when it comes to headphones.
IMG_1075.PNG

I think... I think this is my EQ graph for the FiiO Q5S DAC/AMP after following that chart I showed you guys going after a flat frequency response based on the chart from www.rtings.com for the HD660S
 
Last edited:
May 29, 2020 at 5:12 AM Post #2 of 3
A 'flat' curve (yes, I do realize that's an oxymoron :p) will never sound good.
It's better to EQ closer to the Harman Target Curve and then adjust to your own personal taste.

When EQ'ing, as much as possible only EQ negatively - drag sliders down.
Sliding them up can quickly introduce clipping and distortion.
Unless the EQ has a 'master volume' slider, then you can slide that down negatively to compensate.

Having your own measuring jig is great for tuning with a higher degree of accuracy, but it's really not necessary.
There shouldn't be significant enough unit-variation for anyone to not put enough trust into published graphs.

The problem is that most graphs do not show you the Harman Target Curve along with the headphone's measured curve.
The other problem is that what sounds good to you might not sound good to someone else.
So, sure, there are 'right' and 'wrong' ways of EQ'ing, but as long as it sounds good to you and you're happy, that's the 'right' way.

I would mostly focus on correcting obvious flaws, like high-treble peaks or missing sub-bass and then adjust to taste from there.
Starting with an extreme EQ like the one you're using above can also easily introduce more audible artefacts and problems than it solves.
 
May 29, 2020 at 10:57 PM Post #3 of 3
Hi,

Have you turned off the Bluetooth after swtiching the EQ in order to save the settings? Or turn off the Q5s directly after setting the EQ?

Best regards
A 'flat' curve (yes, I do realize that's an oxymoron :p) will never sound good.
It's better to EQ closer to the Harman Target Curve and then adjust to your own personal taste.

When EQ'ing, as much as possible only EQ negatively - drag sliders down.
Sliding them up can quickly introduce clipping and distortion.
Unless the EQ has a 'master volume' slider, then you can slide that down negatively to compensate.

Having your own measuring jig is great for tuning with a higher degree of accuracy, but it's really not necessary.
There shouldn't be significant enough unit-variation for anyone to not put enough trust into published graphs.

The problem is that most graphs do not show you the Harman Target Curve along with the headphone's measured curve.
The other problem is that what sounds good to you might not sound good to someone else.
So, sure, there are 'right' and 'wrong' ways of EQ'ing, but as long as it sounds good to you and you're happy, that's the 'right' way.

I would mostly focus on correcting obvious flaws, like high-treble peaks or missing sub-bass and then adjust to taste from there.
Starting with an extreme EQ like the one you're using above can also easily introduce more audible artefacts and problems than it solves.
Thank you for the reply, it was helpful. If I can't figure out the Harman Target Curve I'll let you know. As far as subtracting before adding to the chart is very true.
EDIT: https://mehlau.net/audio/dirac-live-2/ I did find this link and I suppose I'll try and figure out a curve for myself!
 
Last edited:

Users who are viewing this thread

Back
Top