[gregorio]

However, Q-factor can also be defined by ratio of stored energy in the system and power loss which is linked to damping/ringing.

How much is the “damping factor” going to be affected?

It doesn't take "very high" Q factor to have "ringing" (mathematically anything over 0.5), but for the ringing to be audibly significant in audio, the Q factor may need to be somewhat high.

What is mathematically possible and what we actually find that’s audible with music signals are two different things.

Theoretically yes, but in practice the signal level in the 2.5 kHz band might be so low that there is no danger of clipping at all.

Sure it can be so low that there’s no danger of clipping and with most music recordings that’s likely to be the case. However, it might also be 1.5dB below peak or less in some cases, which would cause clipping. So what are you suggesting, analyse every track you play to make sure it’s always >1.6dB below peak in the 2.5kHz band or just live with the clipping if/when it occurs?

G

 

[Ghoostknight]

imo ringing of normal IIR eq filters IS audible (or its just the phaseshift that is audible), you guys should atleast try a linear phase EQ

 

[bigshot]

You must be talking about analog equalizers. I've never run into any audible problems with digital EQ (as long as it's applied properly, which with a lot of audiophools isn't guaranteed.)

 

[Ghoostknight]

You must be talking about analog equalizers. I've never run into any audible problems with digital EQ (as long as it's applied properly, which with a lot of audiophools isn't guaranteed.)

arent digital IIR filters replications of analog filters?

corrections from AutoEQ have far higher Q then you typically would do yourself, yes they are mostly fine, and its not hugely audible but imo after some testing its audible and FIR/FFT/SPM filters just sound better then IIR ones, and most standart EQ`s are based on IIR

imo there isnt even more of a difference between low and high Q (beside the obvious bandwidth reduction) but its a general problem with phaseshift imo

 

[bigshot]

How can you tell you aren't just hearing the EQ correction?

 

[Ghoostknight]

How can you tell you aren't just hearing the EQ correction?

1. small adjustments already have an effect on SQ, not just frequency response
2. it got really obvious as soon i used a VST Plugin for eq which lets you switch easly between IIR/FIR/FFT/SPM processing, there is just a difference between those, which wouldnt be there if "anything is inaudible"

 

[bigshot]

I have no idea how you can tell slight differences in phase if the response curves aren't the same. I am betting you didn't do a proper blind test with level matching. The different varieties of EQ probably have slightly different levels.

 

[Ghoostknight]

I have no idea how you can tell slight differences in phase if the response curves aren't the same. I am betting you didn't do a proper blind test with level matching.

for what do i need a blind test if i can easly compare IIR vs FIR with the same response curve??

 

[bigshot]

How do you know they have the same output levels and corrections? Did you run a calibration test with tones? Do they sound different with no response adjustment at all? Do all of them sound different or just one? Which one is the correct one, and on what do you base that?

 

[Ghoostknight]

How do you know they have the same output level?

because the eq i use use the exact same filters for all processing types

Do all of them sound different or just one? Which one is the correct one, and on what do you base that

well SPM sounds a bit different then the others, FIR and FFT sound kinda close (but not exactly the same) but there is a clear difference between all of these and IIR which leaves me to think the phaseshift is the obvious difference

FIR sounds most natural/most accurate to the source SQ-wise in my book, tho SPM is worth a try but it sounds less natural (still better then IIR) imo

 

[bigshot]

Welp! A blind test would verify that's correct! I think you've got some expectation bias going there.

 

[castleofargh]

You also mentioned you raised a band around 2.5k by 5dB and lowered the EQ input by 3.4dB. To be sure to avoid clipping, you should lower the input by at least the same amount as you’re boosting (5dB) and preferably a little more.

He's talking about parametric and graphic visual before, so I'm guessing that other bands lower that same area, and he picked as gain compensation the peak amplitude shown on the EQ curve.
Or maybe I'm completely off ^_^.

imo ringing of normal IIR eq filters IS audible (or its just the phaseshift that is audible), you guys should atleast try a linear phase EQ

Like with anything else, talking about audibility without even a concept of magnitude is never going to get us past silly audiophoolery.

If we're discussing phase shift in the context of using several tracks of mics recording the same instrument(and EQing only some), or EQing only the left channel of our typical song, then even rather small actions can have clear impacts on perceived sound (in some frequencies more than others). But for the generic album listener trying to correct some headphone annoyance for the entire signal and in both channels at once, I'd argue that probably nobody could tell what filter is used just by hearing a result in isolation. Maybe direct A/B could give it away sometimes, but then we'd have to set a listening session so that the delay introduced when switching in one direction isn't what gives away the filter used.

About blind test, anytime you're the guy pushing the buttons, your brain could tell you it's hearing anything you want it to hear. And as now documented in research, there might not even be much of a difference in the electric actions of the auditory cortex between imagining and hearing something(which might be part of why some people are so very adamant about "knowing" what they "heard" even if the test played the very same signal twice.
Anytime differences are small, or mixed with other cues, anytime there are discussion about what's audible, we should be real careful not to rely on sighted impressions and gut feelings to draw conclusions. Nobody follows that advice, not even I most of the time, but it's still good and justified advice for anybody interested in the truth.

 

[Draxu]

Hi all -

I'm relatively new here but I've built a nice little collection of stuff and I want to share what I guess would be considered a semi-controversial observation (I think?):

I hate to EQ.

Every time I apply the EQ presets that are custom-tailored for certain headphones or IEMs, the overall sound experience sounds thinner/less warm/more metallic. The decibel level clearly goes down so maybe some of the difference is attributable to that? On Mac I'm using SoundSource and Qudelix 5K on portable.

I'm not posting this here to pick a fight with EQ-lovers. I'm just curious if I'm doing something wrong?

P.S. I will say there is one case where EQ really felt like it was needed: UM Mest MKII. It was a muddled mess unless I put EQ on. The rest of my stuff (u12t and Arya) felt like they were better without EQ.

Any other insights/tricks I'm missing?

Rich

If you're using your Qudelix 5k it has a great parametric EQ feature. This comes extremely in handy if you know what kind of sound signature you prefer. If you use a graphing tool like squiglink you can automatically generate an eq setting based off another headphone with your preferred sound signature. Also I'd adjust your pre-amp to avoid clipping in cases where you're increasing frequency dB (you can see what the highest dB increase is on bottom of the graph in your app, in case your frequency adjustments affect each other).

 

[71 dB]

How much is the “damping factor” going to be affected?

Affected by what?

What is mathematically possible and what we actually find that’s audible with music signals are two different things.

Of course. Q factor is defined mathematically. When I talk about "ringing" related to Q factor, I talk about mathematical/physical ringing, not audible ringing. These two things are of course related and large enough mathematical ringing is also audible.

Sure it can be so low that there’s no danger of clipping and with most music recordings that’s likely to be the case. However, it might also be 1.5dB below peak or less in some cases, which would cause clipping. So what are you suggesting, analyse every track you play to make sure it’s always >1.6dB below peak in the 2.5kHz band or just live with the clipping if/when it occurs?

G

I am pretty sure my ears would condemn music containing that much energy in the 2.5 kHz band unlistenable. I suggest knowing what you are doing. The signal peaks are constructed from the signal energy on the whole frequency range. If we look at the highest peak, part of it is from 2.5 kHz band. If our music was 10 octaves wide band (20 Hz to 20 kHz) of spectrum that goes down 3 dB/octave (as pink noise does), the each octave would contain 10 % of the whole energy. Of course the peaks in each band are somewhat random, but you would need to increase the level in one octave band massively to compensate the overall level drop of 3.4 dB. The tricky part here is the crest factor of the signal, which makes the math complex.

The point is music doesn't live only in the 2.5 kHz band. Only a fraction of the overall energy lives there. That's why even if your music has peaks of 0 dBFS, the peaks in the 2.5 kHz band are much lower (below -3 dBFS). Dropping overall level by 3.4 dB seems enough safety margin to me.

 

[Fargeg]

-snip-