[71 dB]

My experience differs. With my Toneboosters parametric EQ, in the range 1kHz - 10kHz, there is very appreciable ringing at Q's over 1.3 or so. The audible effect is a sort of smearing and overbrightness especially noticeable on sharp transient sounds like guitar plucks.

Are you talking about notch filters? Q=1.3 is very small value for notch filters. Typically notch filters are used to remove specific non-desired frequencies such as 50 Hz/60 Hz mains hum using very high Q factor (e.g. 25). Notch filter remove the filtered frequency completely. I usually use reduced notches that limit the "pothole" to say -18 dB applying it in parallel mode and mixing the filtered and original signal together in a ratio that gives the desired deepness. For example 25 % original and 75 % filtered gives 12 dB deep hole.

If you were talking about boosts then you are correct as 1.3 starts to be a lot for Q factor in audio.

 

[dmagnan]

Are you talking about notch filters? Q=1.3 is very small value for notch filters. Typically notch filters are used to remove specific non-desired frequencies such as 50 Hz/60 Hz mains hum using very high Q factor (e.g. 25). Notch filter remove the filtered frequency completely. I usually use reduced notches that limit the "pothole" to say -18 dB applying it in parallel mode and mixing the filtered and original signal together in a ratio that gives the desired deepness. For example 25 % original and 75 % filtered gives 12 dB deep hole.

If you were talking about boosts then you are correct as 1.3 starts to be a lot for Q factor in audio.

Yes, I was talking about bell and shelf-type filters. I'm not using any notch filters. I have found that with many or most published tables of recommended parametric EQ values, the ringing associated with the high Q values detracts too much from the sound. So I readjust the Q values down. Frequency response isn't as well corrected, but the overall result is more musical and realistic.

 

[dmagnan]

My experience differs. With my Toneboosters parametric EQ, in the range 1kHz - 10kHz, there is very appreciable ringing at Q's over 1.3 or so. The audible effect is a sort of smearing and overbrightness especially noticeable on sharp transient sounds like guitar plucks.

I should have explained that the filters I am using are not notch filters, but regular bell and shelf types.

 

[71 dB]

Yes, I was talking about bell and shelf-type filters. I'm not using any notch filters. I have found that with many or most published tables of recommended parametric EQ values, the ringing associated with the high Q values detracts too much from the sound. So I readjust the Q values down. Frequency response isn't as well corrected, but the overall result is more musical and realistic.

I should have explained that the filters I am using are not notch filters, but regular bell and shelf types.

What you write here makes total sense, but gregorio was talking about notch filters (for some reason) while mentioning how very high Q values are okay and he is correct about that.

 

[dmagnan]

Are you talking about notch filters? Q=1.3 is very small value for notch filters. Typically notch filters are used to remove specific non-desired frequencies such as 50 Hz/60 Hz mains hum using very high Q factor (e.g. 25). Notch filter remove the filtered frequency completely. I usually use reduced notches that limit the "pothole" to say -18 dB applying it in parallel mode and mixing the filtered and original signal together in a ratio that gives the desired deepness. For example 25 % original and 75 % filtered gives 12 dB deep hole.

If you were talking about boosts then you are correct as 1.3 starts to be a lot for Q factor in audio.

Unfortunately the Toneboosters parametric EQ doesn't have a parallel mode.

 

[gregorio]

Try the same "Q"-settings for narrow band boosts and it is another story!

And why would you use very narrow band boosts to correct headphones?

My experience differs. With my Toneboosters parametric EQ, in the range 1kHz - 10kHz, there is very appreciable ringing at Q's over 1.3 or so.

There shouldn’t be “very appreciable” ringing, unless you’re using high Q factors and boosting significantly, which you shouldn’t be when correcting HPs. Are you sure it’s even ringing? While I only used the Toneboosters EQ a few times and that was several years ago, its “feature” over most other EQs is that it’s a modelled EQ. It emulates a vintage analogue hardware EQ including it’s non-linear response!

but gregorio was talking about notch filters (for some reason)

Because we’re discussing the relationship between high Q settings and audible effects/ringing, and notch filters represent the most extreme Q settings. We’re arguing about Q settings marginally higher than 0.8 but notch filters quite commonly have Q settings of 1,000-4,000 with attenuation >40dB! That’s massively high Q compared to the settings we’re discussing here, yet still there’s rarely any audible ringing. So clearly the “relationship” is not as is being characterised.

G

 

[gregorio]

i have however wondered, generally with most things audio we will prefer analog hardware over software, …

That tends to be true in the audiophile community but less true in wider audio community, especially the pro audio community. This is mainly down to the false marketing so prevalent in the audiophile community and largely stems from Sony’s marketing of SACD.

its pretty rare you see hardware equalizers these days (i remember them being common in the 80s) its always done with software now. is there really no benefit to a hardware EQ vs a software one?

In pro audio, hardware EQs were still prevalent well into the 1990’s and they can still be found in many of the top commercial music studios. There are numerous advantages/benefits to software EQs; cost is an obvious one, high quality hardware EQ will cost hundreds of dollars per channel, so 72 channels is going to be many tens of thousands, while software EQ is at most a couple of hundred bucks for hundreds of channels. In addition, there’s huge functionality benefits, not only many more bands of EQ and many more configurable options but also the recall of settings and changing them dynamically over time (“automation”). However, some of these benefits can at times be hindrances. With 60 or more configurable options, it’s easy to spend several/many minutes tweaking them, while with just 6-8 knobs on a hardware unit there’s only so much you can do and commonly you can do it in a few seconds or so, which means you don’t loose your “train of thought” if you notice something needs an EQ adjustment while working on some other aspect of the mix. There’s also the potential benefit of the buildup of certain desirable artefacts, as VNandor mentioned, although these artefacts aren’t always desirable. The top studios which still maintain these analogue desks/EQs actually record into DAWs (Pro Tools) which means producers/engineers can use either hardware or software EQs, or even both at the same time.

For consumers, the only benefits of analogue hardware units I can think of are cosmetic (they like the look of an EQ unit in their setup) and/or “bragging rights” to others who don’t know any better or are similarly deluded by the false audiophile marketing.

G

 

[71 dB]

And why would you use very narrow band boosts to correct headphones?

Because we’re discussing the relationship between high Q settings and audible effects/ringing, and notch filters represent the most extreme Q settings. We’re arguing about Q settings marginally higher than 0.8 but notch filters quite commonly have Q settings of 1,000-4,000 with attenuation >40dB! That’s massively high Q compared to the settings we’re discussing here, yet still there’s rarely any audible ringing. So clearly the “relationship” is not as is being characterised.

G

And why would you use notch filters to correct headphones? Yeah, this discussion has lost all of its meaningfulness at this point. I kind of regret I ever participated. I haven't been talking only about correcting headphones because filters can be used elsewhere too where they have "job to do." I don't like when discussion forums go this way and it happens often because there are always people who want the last word to look the smartest guy. Maybe my presence is not needed here and I can waste my time elsewhere because gregorio knows best...

 

[gregorio]

And why would you use notch filters to correct headphones?

You wouldn’t but I just used notch filters to demonstrate that even extremely/massively high Q settings don’t necessarily cause audible ringing.

I haven't been talking only about correcting headphones because filters can be used elsewhere too where they have "job to do."

But I was talking only about correcting HPs and it was that post you decided needed correcting/“refining”.

I don't like when discussion forums go this way and it happens often because there are always people who want the last word to look the smartest guy of all.

You’ve “corrected” some of my posts in the past and on occasion you were right to do so, other times you usefully expanded on what I stated within the context of the discussion but a few times recently you’ve decided to “correct” my posts falsely/outside the context of the discussion, which is starting to seem like an agenda. If you “don’t like when discussion forums go this way”, then why are you going this way?

Maybe my presence is not needed here and I can waste my time elsewhere.

Your contributions are often useful/valuable, in fact, you posting the mathematical proofs can be invaluable but on occasion of course, the maths can far exceed what we can hear or it’s application maybe somewhat different in practice.

G

 

[emba4]

That tends to be true in the audiophile community but less true in wider audio community, especially the pro audio community. This is mainly down to the false marketing so prevalent in the audiophile community and largely stems from Sony’s marketing of SACD.

In pro audio, hardware EQs were still prevalent well into the 1990’s and they can still be found in many of the top commercial music studios. There are numerous advantages/benefits to software EQs; cost is an obvious one, high quality hardware EQ will cost hundreds of dollars per channel, so 72 channels is going to be many tens of thousands, while software EQ is at most a couple of hundred bucks for hundreds of channels. In addition, there’s huge functionality benefits, not only many more bands of EQ and many more configurable options but also the recall of settings and changing them dynamically over time (“automation”). However, some of these benefits can at times be hindrances. With 60 or more configurable options, it’s easy to spend several/many minutes tweaking them, while with just 6-8 knobs on a hardware unit there’s only so much you can do and commonly you can do it in a few seconds or so, which means you don’t loose your “train of thought” if you notice something needs an EQ adjustment while working on some other aspect of the mix. There’s also the potential benefit of the buildup of certain desirable artefacts, as VNandor mentioned, although these artefacts aren’t always desirable. The top studios which still maintain these analogue desks/EQs actually record into DAWs (Pro Tools) which means producers/engineers can use either hardware or software EQs, or even both at the same time.

For consumers, the only benefits of analogue hardware units I can think of are cosmetic (they like the look of an EQ unit in their setup) and/or “bragging rights” to others who don’t know any better or are similarly deluded by the false audiophile marketing.

G

thank you, i love the replies i get here, i learn stuff

im not gonna lie cheesy as it may be i would kinda like a big flashy 80s graphic equalizer, but pretty much just for the look lol

 

[dmagnan]

That tends to be true in the audiophile community but less true in wider audio community, especially the pro audio community. This is mainly down to the false marketing so prevalent in the audiophile community and largely stems from Sony’s marketing of SACD.

In pro audio, hardware EQs were still prevalent well into the 1990’s and they can still be found in many of the top commercial music studios. There are numerous advantages/benefits to software EQs; cost is an obvious one, high quality hardware EQ will cost hundreds of dollars per channel, so 72 channels is going to be many tens of thousands, while software EQ is at most a couple of hundred bucks for hundreds of channels. In addition, there’s huge functionality benefits, not only many more bands of EQ and many more configurable options but also the recall of settings and changing them dynamically over time (“automation”). However, some of these benefits can at times be hindrances. With 60 or more configurable options, it’s easy to spend several/many minutes tweaking them, while with just 6-8 knobs on a hardware unit there’s only so much you can do and commonly you can do it in a few seconds or so, which means you don’t loose your “train of thought” if you notice something needs an EQ adjustment while working on some other aspect of the mix. There’s also the potential benefit of the buildup of certain desirable artefacts, as VNandor mentioned, although these artefacts aren’t always desirable. The top studios which still maintain these analogue desks/EQs actually record into DAWs (Pro Tools) which means producers/engineers can use either hardware or software EQs, or even both at the same time.

For consumers, the only benefits of analogue hardware units I can think of are cosmetic (they like the look of an EQ unit in their setup) and/or “bragging rights” to others who don’t know any better or are similarly deluded by the false audiophile marketing.

G

I think that there are persistent sonic differences between hardware RLC tuned circuit equalizers and software equalizers. They use completely different implementations, and have different pros and cons. Digital processing totally macerates the sound to change it into another effective mathematical dimension of sample time space in which there is considerable musical information loss especially at high frequencies, due to the PCM sampling rates and the bit depth. The music is now a mathematical approximation with certain data losses and distortion additions like aliasing. The digital system then macerates the music again to manipulate the numbers and bits to implement filter algorithms, then transforms the bit data back into analog in the DAC. All this processing takes its toll of versimilitude to live music - losses in resolution and additions of digital "grunge" especially in the high frequencies in such ways that especially effect naturalness of massed strings, due to the great sensitivity of the human ear to time-related distortions.

Whereas the analog EQ system keeps the dimensional system of the music the same, and all processes are analog , continuous in time, and don't operate mathematically on chopped-up data-lossed replicas of the recorded music. But at the same time, the analog system has much less capability to implement various types of filters, is subject to various analog distortions due to nonlinearities in the active transistor and tube devices and in the passive RLC components. And they lose a little resolution due to the analog noise floor being usually much higher than the digital one. But because of the relative lack of time related distortions and other digital artifacts, to me the tradeoff is in favor of the analog EQ. Easier on the ear.

 

[doesthislookinfected]

In my Yamaha AVR two types of frequency set up – PEQ & GEQ
Band are as follows :-
63Hz
160Hz
400Hz
1KHz
2.5KHz
6.3KHz
16KHz
Please suggest which one is better GEQ or PEQ
Best set up in above band for both movies and music?

 

[Ghoostknight]

Please suggest which one is better GEQ or PEQ
Best set up in above band for both movies and music?

PEQ is always superior to fixed band GEQ, GEQ might get the job somewhat ok done but PEQ offers way more adjustments, including other filter types then bell

 

[gregorio]

Digital processing totally macerates the sound to change it into another effective mathematical dimension of sample time space in which there is considerable musical information loss especially at high frequencies, due to the PCM sampling rates and the bit depth.

You have this backwards. There is no musical information loss, especially in the high frequencies, or at least none anywhere near audible levels within the audible spectrum, due to the sampling rates and bit depth. This is in contrast to analogue with all kinds of tape hiss, distortion, saturation and other “macerations” at high freqs which are audible!

The music is now a mathematical approximation with certain data losses and distortion additions like aliasing.

Again, no. It’s a highly accurate mathematical representation with no data losses, distortions or aliasing that are audible.

The digital system then macerates the music again to manipulate the numbers and bits to implement filter algorithms,

The digital system has not macerated the music in the first place and does not macerate it again. Again, this is in contrast to analogue, which does add distortion, noise, etc., in the first place and adds more with each subsequent process.

All this processing takes its toll of versimilitude to live music

No it doesn’t, the “toll” occurs at bit depths way beyond anything that can even be reproduced downstream, let alone at audible levels, again, unlike analogue.

- losses in resolution and additions of digital "grunge" especially in the high frequencies in such ways that especially effect naturalness of massed strings, due to the great sensitivity of the human ear to time-related distortions.

There is no loss in resolution and no such thing as “digital grunge”, especially in the high frequencies and therefore it does not affect the naturalness of massed strings. And, the human ear does not have “great sensitivity to time-related distortions” compared to the time related distortions which actually occur with digital audio. If the human ear were so sensitive, you would easily hear all the time-related distortions of analogue, which are magnitudes greater!

Whereas the analog EQ system keeps the dimensional system of the music the same,

Of course it doesn’t, music exists in the acoustic “dimension” NOT the analogue “dimension”. That’s why we need microphones to change the “dimension” and speakers/HPs to change it back again. Surely you must know this?

and all processes are analog , continuous in time, and don't operate mathematically on chopped-up data-lossed replicas of the recorded music.

Of course they operate mathematically, EQ and other processes are defined mathematically and then circuits designed to implement that math, even the pick-up patterns/characteristics of mics. How do you think EQ and other analogue processes/processors operate?

But at the same time, the analog system has much less capability to implement various types of filters, is subject to various analog distortions due to nonlinearities in the active transistor and tube devices and in the passive RLC components.

Now you’re contradicting yourself. Analogue has all these and other distortions, which digital either doesn’t have at all or has magnitudes lower but digital “macerates” the music and analogue doesn’t?

But because of the relative lack of time related distortions and other digital artifacts, to me the tradeoff is in favor of the analog EQ.

What relative lack of time related distortions? Analogue EQ has exactly the same mathematic processes as digital EQ processors, with exactly the same artefacts plus additional noise and distortion, while digital does not have “other digital artefacts”! Where’s the trade off?

Easier on the ear.

You mean you prefer the higher noise, distortion and other “macerations” of analogue!

Unfortunately, your post was packed with a whole list of the false marketing invented and employed by the audiophile community since Sony’s infamous SACD/DSD marketing. It’s completely contrary to the actual facts/science and is based on myth and falsehoods! Although it did elucidate exactly the false audiophile marketing I mentioned.

G

 

[Ghoostknight]

I think that there are persistent sonic differences between hardware RLC tuned circuit equalizers and software equalizers. They use completely different implementations, and have different pros and cons. Digital processing totally macerates the sound to change it into another effective mathematical dimension of sample time space in which there is considerable musical information loss especially at high frequencies, due to the PCM sampling rates and the bit depth. The music is now a mathematical approximation with certain data losses and distortion additions like aliasing. The digital system then macerates the music again to manipulate the numbers and bits to implement filter algorithms, then transforms the bit data back into analog in the DAC. All this processing takes its toll of versimilitude to live music - losses in resolution and additions of digital "grunge" especially in the high frequencies in such ways that especially effect naturalness of massed strings, due to the great sensitivity of the human ear to time-related distortions.

the biggest sonic differences i noticed with EQ`s are that nearly all digital eq`s sound indeed different BUT because they implement filters in a similar way to analog eq, including phase shift, FIR (LINEAR unaltered phase!) processing sounds way superior to good old IIR filters which most EQ`s use

if you actually prefer RLC filters "Linux Studio Plugins - Parametric Equalizer" VST plugin actually offer them, including completely digital filters in APO style
this plugin also lets you process data in IIR / FIR / FFT and SPM, worth a try, FIR and SPM sounds best to me