good point
hmm ... maybe it does cause some 'fluctuation'
music is dynamic, but EQ is static
lets suggest you change the midband response
it is obvious how it will influence the pass band response
but it is much less obvious what happens in the 'stopband'
(expecting knowledge about how crossovers work)
Looking at magnitude response plots before and after EQ always looks impressive but those plots can be highly misleading. They show how "loud" each frequency is represented within a certain time frame.
1. This is not how our hearing works. We do perceive sound in time. Reflections of clicks might cause audible echoes when they arrive within a few milliseconds, whereas reflections of a violin will cause audible echo only when the delay is very long. The magnitude response is completely blind to those effects.
2. They can vary a lot with position. If we base correction only on a single point then other points might get worse. Humans have two ears separated by about 0.5 ft and we don't tend to sit still while listening. So we would need to look at multiple points within an area about 0.5 ft deep/high and about 1 ft wide. In a home theater the area is even larger.
The best description on the topic in layman's terms I've found so far can be found in the PDF linked in post #9:
Consider a loudspeaker standing in a room. Mr A measures impulse responses in a certain listening volume and finds to his dismay that the magnitude response has a substantial broad dip at some rather low frequency, say 300 Hz, in all positions. He calibrates a peak filter and fills up the hole in the magnitude response, which is then confirmed by measurements. Enter Mr B. Mr B is a musician and he listens to the equalized system. “It sounds horrible! What have you done to the system!? It sounds all swollen and strange!” Mr A becomes nervous, as Mr B is an important customer, and calls his trusted friend Mr C. Mr C answers: “Ah, yes of course. The dip was really due to reflections. You should never boost any dip, because they are typically due to reflections.” So Mr A removes his equalizer filter and lets Mr B listen again. Mr B, however, is still not happy. “It is better, but it’s not good. There is something hollow about the sound.” At this time Mrs D enters the conversation. She’s been listening, sitting quietly in a corner of the room, and says: “Mr A was wrong because he forgot about the time domain. Looking only at the magnitude of the Fourier transform and interpreting it as strongly related to our concept of frequencies, he thought that he could boost that region and obtain better sound. The problem is that he uses minimum-phase filters and consequently adds energy at that frequency early in time. But if we only look at the direct wave there is no hole to be filled in the frequency response. The hole never exists if we look at a short window at any time.” Mr B frowns: “So Mr C was right to say that we cannot do anything about it. But if that’s the case, why do I still hear a strange sounding oboe on my recording?” Mrs D looks sternly at him: ”Mr C was wrong too. The problem is due to the time domain properties; the reflection causes the problem and it can only be corrected for by a time-domain approach. If we design a filter that reduces the reflection, you will end up with the interesting result that the hole will be gone and the oboe will sound more natural.” “But,” Mrs D adds, “don’t take this example as evidence that you can always correct dips this way! In this case it was possible, because all positions experienced the same problem.”
Last edited:
Hi Markus,
The Yamaha YDP2006 is a good Digital Parametric Equalizer ($2,300 retail) you can get for $200 used.
=> YDP2006 - Processors - Live Sound - Products - Yamaha United States
* Inexpensive way to tailor your preferred sound into your ears. ...Room/Speaker correction, all that acoustic jazz.
The Yamaha YDP2006 is a good Digital Parametric Equalizer ($2,300 retail) you can get for $200 used.
=> YDP2006 - Processors - Live Sound - Products - Yamaha United States
* Inexpensive way to tailor your preferred sound into your ears. ...Room/Speaker correction, all that acoustic jazz.
Last edited:
good point
hmm ... maybe it does cause some 'fluctuation'
music is dynamic, but EQ is static
lets suggest you change the midband response
it is obvious how it will influence the pass band response
but it is much less obvious what happens in the 'stopband'
(expecting knowledge about how crossovers work)
A crossover represents a linear filter. It doesn't change.
Be patient, we are slowly but surely coming. ...We don't want to miss a single beat.
Last edited:
A mixture of room treatments and judicious "softening" EQ. ...Peaks, not dips.
You can also use a simple Treble tone control to give life to those dead albums and CDs (by adding 1.5-2.5dBs),
and by respecting the master volume level (not going to eleven).
Yeah? . ..No? ...Of course yes.
Last edited:
Just a few comments from me, I think I wrote them down here already a while ago:
- trying to equalize your room modes with FIRs is... stupid, to be polite. You'll need very long filters to make it work. Using IIRs is a much better approach
- above the Schroeder frequency there's nothing bad about FIRs, also PEQs/GEQs still work well (but designing an FIR EQ is so simple!)
- trying to equalize room modes over a wide room area doesn't work, when you have only one subwoofer; to have a flat bass response over a wide area you are forced to use a lot of damping or one of the many multi bass approaches; so measuring at one point in space is enough (probably measuring at different locations can lead to a good compromise)
- that said, it is not necessary to use such an optimising approach as linked above, and probably unpractical if you try to implement that on a small and cheap integer only µC where you're already out of memory. Using some sort of "best guess" is sufficient and works well here.
- above Schroeder frequency you can use multiple meausrements point and average them before you design the equalizers
In general, I don't here any big differences between FIR or PEQ equalizing. The advantage of FIR is the simple design of them, the advantage of PEQ is the cheap implementation (computation time and memory).
- trying to equalize your room modes with FIRs is... stupid, to be polite. You'll need very long filters to make it work. Using IIRs is a much better approach
- above the Schroeder frequency there's nothing bad about FIRs, also PEQs/GEQs still work well (but designing an FIR EQ is so simple!)
- trying to equalize room modes over a wide room area doesn't work, when you have only one subwoofer; to have a flat bass response over a wide area you are forced to use a lot of damping or one of the many multi bass approaches; so measuring at one point in space is enough (probably measuring at different locations can lead to a good compromise)
- that said, it is not necessary to use such an optimising approach as linked above, and probably unpractical if you try to implement that on a small and cheap integer only µC where you're already out of memory. Using some sort of "best guess" is sufficient and works well here.
- above Schroeder frequency you can use multiple meausrements point and average them before you design the equalizers
In general, I don't here any big differences between FIR or PEQ equalizing. The advantage of FIR is the simple design of them, the advantage of PEQ is the cheap implementation (computation time and memory).
only trying to look at why 'simple' EQ might have some less expected effects
cheers
From what I've gathered from reading some other posts and papers on the subject, it seems that we run into problems as soon as we go from "equalize using a short series of single tones or a quick sweep" to "equalize for music", meaning once we have longer duration sounds playing and multiple sounds / harmonics, things start interacting in such a way that our eq settings that we used to correct a "simple signal" start to fall apart, as in an earlier post, "why does this oboe sound like crap when we've supposedly EQ'd those bands neutral"
I've often wondered in this age of super fast technology, why we don't use active correction if at all, I.E similar to a servo in a subwoofer, but extended to the whole room using microphones and processing.
Think "noise cancelling headphones" technology, but applied to an entire room. Couldn't that work? The room just "eats" everything that's not in the recording. I'd imagine it would be funny to try having a conversation in such a room, but who knows, it might be magic for audio reproduction..and we've already got all the tech in place.
True, there's not much DIY potential in computer based systems, unless you're the programmer that comes up with the methods, but we could really say the same thing about any computer based tool that we use for anything audio related.
Then again, if it's just a box that has 20 microphone jacks and a series of audio I/O plugs, it'd still be up to us to experiment and measure, the same as how we'd use REW or anything else currently.
Maybe I need to start learning how to code
Actually, it is how our hearing works. Have a look at JJ Johnston's research on psychoacoustics: PowerPoint Presentations from recent (or not so recent) meetings. JJ is considered the expert on how our hearing works.
Frequency dependent windowing (FDW) is the key to analyzing room acoustics and very few acoustic analysis software has this capability. Those magnitude plots use FDW where the window is long, 100's of ms at low frequency and short at high frequencies, <1 ms.
The Dirac Live paper is good marketing for sure. As an ex recording/missing engineer, I have lived this exact problem moving from one studio to the next. However, most studios had at least monitor system eq'd to the B&K curve, which is very similar to the charts I posted. The B&K curve was figured out in the 70's for that vary reason. Not much has changed since
Re: vary in position is an old audiophile wives tale. Again FDW comes into play as the FIR correction at low frequencies is in the 100's of ms. Think about what that means as one moves from the center of the listening position to one side of a couch or another or anywhere in the room for that matter. I.e. convert the time based correction (i.e. 100's of ms) into distance.
Cheers, Mitch
Well, 100ms is around 34.3m, of course, but generally any response problem at low frequencies that would vary with position would be from room modes so... I don't understand your point? What's the old wives' tale?
The FIR correction filter is correcting magnitude response over time. At 40 Hz., depending on filter length, may have 400ms of correction over the bass frequencies. So... the bass sounds even in magnitude response over time. Converting to distance, this is why moving around the couch, the bass still sounds even.
It is not just one point in time, it is over the FDW time. The olds wives tale is that if you move your head by 0.5 ft the bass response is very different. Maybe the case with traditional eq, like PEQ that is not time based. People should try and hear for themselves.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.