Paul Hales on room equalization on Soundstage! Network

[goat76]

Yes, I am a strong advocate of using ears. And I also agree that completely flattening peaks often does not yield the best result.

And that is at least an indication that it may be some truth in what Paul Hales says in the podcast.

 

[SigbergAudio]

And that is at least an indication that it may be some truth in what Paul Hales says in the podcast.

Is it possible to summarize his point or refer to somewhere in the podcast it is summarized, so one doesn't have to listen to 44 minutes?

 

[Vuki]

Is it possible to summarize his point or refer to somewhere in the podcast it is summarized, so one doesn't have to listen to 44 minutes?

By applying room eq one is also altering direct sound making it less linear which is noticable by listening.

 

[singular]

The EQ by Schroeder is very easy and old technique - everybody can find that as HP on old style equalizers.

The problem is always how the speakers ( bass-reflex system ) is made and how excessive is their GD.
The same problem could be with subwoofers and their different phase and GD.
Trying to match mains to subs, when they have different phase, could lead to bass pursuit.
So is easier to go FIR than change crossover.

Btw these old analog equalizers do have everyrhing needed to match well made speakers to room.

If the speakers are made to match the room fundamental, there is almost no problem with reverb time at frequencies longer than room fundamental - below Schroeder.

Also some speakers have first impedance peak very low - this tend to reverb subbas in smaller apartment fundamental.
You can add passive 1st order HP in signal path and the result will be even better above this frequency, without adding GD.

Bigger problem which is audible and can destroy scene, attack or impact, and nature of music, is High Q equalised Midrange.

In the physical room there are not higher Q than certain value and level, so Midrange could be equalised in those values for frequencies shorter than room fundamental.

Less is more here.

Dips and Peaks are always products of untreated room and tend to migrate in frequency until the room is equaly treated on all surfaces - but this always happen on frequencies above Schroeder - whole midrange and trebles are affected by this.

So my opinion is that using FIR and subwoofers, to equalize the mains also below Shroeder, is just easy way how to overcome physical variables of untreated room, and differences between mains and subs.

 

[SigbergAudio]

By applying room eq one is also altering direct sound making it less linear which is noticable by listening.

Finally listened to most of it. Feels like he's got a few fundamentals wrong, and is mixing up what's happening below vs above the transition frequency of the room.

1. He keeps talking about better loudspeakers. This is an argument above the transition frequency, but in the bass even the objectively best speaker in the world will still excite room modes.
2. He is mixing original sound with reproduced sound. The recording already has room information on it (artificial or real), we don't necessarily want our own room to add too much to it as well. This is somewhat of a sidenote as we want that solved by acoustic treatment rather than EQ.
3. He's arguing that EQ doesn't sound natural, I'd agree in the higher frequencies, but certainly not in the bass (again done right).
4. I suspect his piano analogy is questionable at best. First of all I think it would radiate quite differently than a speaker, second of all I think less of what is emitted from a piano is in the sub 100hz range than we think. Even from the lowest notes, it's not like the piano is pushing out the fundamental very loud, we hear mostly harmonics. So most of the piano actually isn't in the range we'd typically EQ.
5. He is apparently sceptical to automatic room EQ, mostly rightly so. But that doesn't invalidate bass EQ.
6. He claims the microphone can't distinguish between reflected and direct sound, while we can. This is only true above the bass, not in the range we are suggesting to EQ.
7. As a follow-up to 6, he claims that this missing differentation between reflected and direct sound is what makes us lose impact. But that argument builds on the false premise of 6. But I will him give him half-right on this, because this is why we don't flatten peaks 100%. We lose impact if we EQ too much.
8. He points out we can't EQ individual seats individually, which is true. But not really an argument against making the sound better in the main listening position.
9. He talks about short transients, but any 50hz note (long or short) will excite the room mode at 50hz if there is one.
10. Why are we completely ignoring all the rest of the sounds in the bass range? We don't listen to just kick drums. He seems to completely ignore the fact that the level of any note that hits a room mode is simply too loud. He claims this sounds more natural, while in reality it does not.

 

[Cote Dazur]

Finally listened to most of it. Feels like he's got a few fundamentals wrong, and is mixing up what's happening below vs above the transition frequency of the room.

1. He keeps talking about better loudspeakers. This is an argument above the transition frequency, but in the bass even the objectively best speaker in the world will still excite room modes.
2. He is mixing original sound with reproduced sound. The recording already has room information on it (artificial or real), we don't necessarily want our own room to add too much to it as well. This is somewhat of a sidenote as we want that solved by acoustic treatment rather than EQ.
3. He's arguing that EQ doesn't sound natural, I'd agree in the higher frequencies, but certainly not in the bass (again done right).
4. I suspect his piano analogy is questionable at best. First of all I think it would radiate quite differently than a speaker, second of all I think less of what is emitted from a piano is in the sub 100hz range than we think. Even from the lowest notes, it's not like the piano is pushing out the fundamental very loud, we hear mostly harmonics. So most of the piano actually isn't in the range we'd typically EQ.
5. He is apparently sceptical to automatic room EQ, mostly rightly so. But that doesn't invalidate bass EQ.
6. He claims the microphone can't distinguish between reflected and direct sound, while we can. This is only true above the bass, not in the range we are suggesting to EQ.
7. As a follow-up to 6, he claims that this missing differentation between reflected and direct sound is what makes us lose impact. But that argument builds on the false premise of 6. But I will him give him half-right on this, because this is why we don't flatten peaks 100%. We lose impact if we EQ too much.
8. He points out we can't EQ individual seats individually, which is true. But not really an argument against making the sound better in the main listening position.
9. He talks about short transients, but any 50hz note (long or short) will excite the room mode at 50hz if there is one.
10. Why are we completely ignoring all the rest of the sounds in the bass range? We don't listen to just kick drums. He seems to completely ignore the fact that the level of any note that hits a room mode is simply too loud. He claims this sounds more natural, while in reality it does not.

great post, as always and great resume, thank you.

 

[kemmler3D]

Finally listened to most of it. Feels like he's got a few fundamentals wrong, and is mixing up what's happening below vs above the transition frequency of the room.

1. He keeps talking about better loudspeakers. This is an argument above the transition frequency, but in the bass even the objectively best speaker in the world will still excite room modes.
2. He is mixing original sound with reproduced sound. The recording already has room information on it (artificial or real), we don't necessarily want our own room to add too much to it as well. This is somewhat of a sidenote as we want that solved by acoustic treatment rather than EQ.
3. He's arguing that EQ doesn't sound natural, I'd agree in the higher frequencies, but certainly not in the bass (again done right).
4. I suspect his piano analogy is questionable at best. First of all I think it would radiate quite differently than a speaker, second of all I think less of what is emitted from a piano is in the sub 100hz range than we think. Even from the lowest notes, it's not like the piano is pushing out the fundamental very loud, we hear mostly harmonics. So most of the piano actually isn't in the range we'd typically EQ.
5. He is apparently sceptical to automatic room EQ, mostly rightly so. But that doesn't invalidate bass EQ.
6. He claims the microphone can't distinguish between reflected and direct sound, while we can. This is only true above the bass, not in the range we are suggesting to EQ.
7. As a follow-up to 6, he claims that this missing differentation between reflected and direct sound is what makes us lose impact. But that argument builds on the false premise of 6. But I will him give him half-right on this, because this is why we don't flatten peaks 100%. We lose impact if we EQ too much.
8. He points out we can't EQ individual seats individually, which is true. But not really an argument against making the sound better in the main listening position.
9. He talks about short transients, but any 50hz note (long or short) will excite the room mode at 50hz if there is one.
10. Why are we completely ignoring all the rest of the sounds in the bass range? We don't listen to just kick drums. He seems to completely ignore the fact that the level of any note that hits a room mode is simply too loud. He claims this sounds more natural, while in reality it does not.

Solid post, thanks for actually watching it, haha.

I would just add that bass drums don't sound like bass drums mostly because of the lowest / fundamental frequency.

If you want to synthesize a bass drum hit, you take a low sine wave (say 50hz) and you modulate the frequency with a very quick exponential decay, so that the frequency might start around 200Hz and drop to (say) 70hz in 100ms or so, then setting down to 50hz much more slowly from there.

So I agree the discussion is overly focused on bass drums, I also think the role of the lowest frequencies is being overstated even there. You can't get a bass drum sound with just low bass frequencies.

 

[Coltr@ne]

Equalizing properly

What exactly do you mean—using a parametric EQ in real time? Or something else?

 

[SigbergAudio]

What exactly do you mean—using a parametric EQ in real time? Or something else?

Equalizing mainly below 100-150hz, targeting the main peaks with parametric EQ, doing so somewhat conservatively, avoiding aggressive ruler-flat correction. Using your ears along the way. 2-4 PEQ points often goes a long way.

 

[gnarly]

Equalizing mainly below 100-150hz, targeting the main peaks with parametric EQ, doing so somewhat conservatively, avoiding aggressive ruler-flat correction. Using your ears along the way. 2-4 PEQ points often goes a long way.

Yes, I don't think anyone with a minimal amount of experience would argue against that.
It's what i do, and I remember Paul Hales saying somewhere in the interview, that the smaller the acoustic space, the more likely grabbling a few room EQs can help.
Even this simple, non-overkill, non-automatic DSP room correction, DOES have an impact on direct sound.
How can it not? You can't turn down the offending modal frequency, without turning down its direct response too.

(Please, no one chime in here and say oh but it can be done with FIR....i'm talking about the very common usage of IIR EQs to address room modes as being discussed.) NOT talking DSP room correction

REW has very good tone burst generator. I like to play low frequency sine bursts of a few cycles though a system with the room EQs both on and off.
Easy to do, and not hard to hear the impact on transients. Sync the bursts with REW's scope, and compare the stimulus burst to a mic capture, if you're measurement inclined.

 

[SigbergAudio]

Yes, I don't think anyone with a minimal amount of experience would argue against that.
It's what i do, and I remember Paul Hales saying somewhere in the interview, that the smaller the acoustic space, the more likely grabbling a few room EQs can help.
Even this simple, non-overkill, non-automatic DSP room correction, DOES have an impact on direct sound.
How can it not? You can't turn down the offending modal frequency, without turning down its direct response too.

I already explained at least twice why this doesn't make sense. The answer is still the same even though you repeat the point.

 

[gnarly]

I already explained at least twice why this doesn't make sense. The answer is still the same even though you repeat the point.

We will just have to agree to disagree.
Or you show me a transfer function of some minimum-phase / IIR EQs used for room modal correction, that does not alter direct sound. Good luck !

 

[goat76]

We can debate which frequency is the source of the perception of impact, broad band vs narrow band, etc, but I am not sure it is central to the point.

There is definitely no sense of ‘bass impact’ down at 40 Hz, but at 60 Hz and probably up to around 80-100 Hz there is an obvious lack of impact if you take out some energy in those frequency areas.

It's all quite easy to test; just fire up a track with a punchy kick drum in a DAW where it's easy to turn on/off a a EQ filter with a 6 dB cut (6-7 Q), and I think you will notice how the ‘bass impact’ is affected by that. Down at 40 Hz, there isn't much impact going on compared to 60 Hz.

 

[SigbergAudio]

There is definitely no sense of ‘bass impact’ down at 40 Hz, but at 60 Hz and probably up to around 80-100 Hz there is an obvious lack of impact if you take out some energy in those frequency areas.

It's all quite easy to test; just fire up a track with a punchy kick drum in a DAW where it's easy to turn on/off a a EQ filter with a 6 dB cut (6-7 Q), and I think you will notice how the ‘bass impact’ is affected by that. Down at 40 Hz, there isn't much impact going on compared to 60 Hz.

I would argue that the lack of impact is simply a reduction in volume. It is no surprise that removing 6dB translates into less impact. I don't think anyone would try to argue against that.

 

[singular]

We will just have to agree to disagree.
Or you show me a transfer function of some minimum-phase / IIR EQs used for room modal correction, that does not alter direct sound. Good luck !

For example almost every 3ways speakers with midrange unit ( not midbass ) - as system -will do a job of shifting phase below 100Hz to bounce to horizontal surfaces.

Btw this can help with room modes.

To make a system with directed bass range you need directed cardioid, and even then it is on bass earlier bounced by horizontal surfaces and then by nearest walls.
Earlier then a bass wave could make amount of cycles in time to be perceived.

Or you have anechoic room, or we talk about midrange direct sound.

 

[singular]

There is definitely no sense of ‘bass impact’ down at 40 Hz, but at 60 Hz and probably up to around 80-100 Hz there is an obvious lack of impact if you take out some energy in those frequency areas.

It's all quite easy to test; just fire up a track with a punchy kick drum in a DAW where it's easy to turn on/off a a EQ filter with a 6 dB cut (6-7 Q), and I think you will notice how the ‘bass impact’ is affected by that. Down at 40 Hz, there isn't much impact going on compared to 60 Hz.

6dB cut by 6-7Q will ruin every range.

 

[goat76]

I would argue that the lack of impact is simply a reduction in volume. It is no surprise that removing 6dB translates into less impact. I don't think anyone would try to argue against that.

Yes, but if you really take your time concentrating on the bass impact and the direct power, the punch that comes from the initial kick instead of just the “roomy all around you” bass sensation that doesn't feel like it comes with the direct sound from the loudspeakers themselves, I think you will notice the difference talked about in the podcast.

 

[SigbergAudio]

We will just have to agree to disagree.
Or you show me a transfer function of some minimum-phase / IIR EQs used for room modal correction, that does not alter direct sound. Good luck !

This is a false premise, as we can't distinguish between the direct and reflected sound at low frequencies. You are making the same mistake as the guy in the video, using logic that is applicable above the transition frequency, below the transition frequency.

The room actually contributes to a lot of the low-end sound level, given the fact that the bass is omnidirectional. Room gain results in a bass level that is significantly higher than the direct signal. The speaker and the room becomes one system where the bass boost from the room is an integral part of the signal. So reducing a mode with EQ isn't subtracting from a correct direct signal, it was never correct to begin with. If we only heard the direct signal by itself in the bass, it would be way too low. So we are correcting the response of the room-speaker-system (as we should) back to where it should have been in the first place.

 

[SigbergAudio]

Yes, but if you really take your time concentrating on the bass impact and the direct power, the punch that comes from the initial kick instead of just the “roomy all around you” bass sensation that doesn't feel like it comes with the direct sound from the loudspeakers themselves, I think you will notice the difference talked about in the podcast.

See my previous reply to Gnarly.

 

[SigbergAudio]

Perhaps it's easier if we get practical. We can use our awesome PEQ assistant.

Let's start with an easy one. Here I have plotted a real response from a real room (red), and we also see the suggested new response (made possible by one PEQ point at 46hz).

Would you guys suggest that this system would sound better without this EQ adjustment?