Waterfalls

[PeteD]

Hi Brucek.

You were right about the settings, but the SPL still look different. Maybe I am missing something else?

Here they both are with right window at 300ms (the target line has been manually adjusted, so it may not be the same):





What setting is recommended for adjusting filters?

Thanks,
Pete

[JohnM]

PeteD wrote:

Regarding number 1, please look at the graphs below. These show filtered response from 10 to 200 Hz on the same scale. I generated the LF waterfall from the frequency response. Why do the SPL in waterfall look higher?

Because you are not allowing for the perspective of the waterfall plot. The SPL axis markings on the LHS of the graph are for the frontmost slice of the waterfall, if you look at the lines on the back wall of the waterfall you can see the 96 and 88dB lines and the rearmost slice lines up between them as it should. You can also use the slice slider to move the rearmost slice to the front to more easily see the actual levels, or change the perspective settings via the "More waterfall controls..." panel.

PeteD wrote:

Below what frequency are you less worried about resonance?

I don't understand that question. Modal resonances are most significant at low frequencies, below about 200Hz. At higher frequencies they are still present but there are so many their individual effects are no longer easily resolved and dealing with them requires room treatment.

[brucek]

Maybe I am missing something else?

As John says, move the 1-30 slider on the Waterfall plot over to 1 and you will have identically response charts.....

Perhaps best you start a new thread in the REW section to discuss your filters etc..... it will get more play there..

brucek

[PeteD]

JohnM wrote:

Because you are not allowing for the perspective of the waterfall plot.

I knew it had to something very simple!

[PeteD]

JohnM wrote:

I don't understand that question. Modal resonances are most significant at low frequencies, below about 200Hz. At higher frequencies they are still present but there are so many their individual effects are no longer easily resolved and dealing with them requires room treatment.

Sorry for the very poorly worded question. I was looking for you to expound in this statement:

"I'd agree that a very "dry" (fast decay time) room is not good for music, though it can work well for movies. I also agree that for low frequency behaviour RT60 measurements are meaningless."

How do we use these waterfall plots, especially at low frequencies if the RT60 become meaningless?

Also, regarding your point on rapid decay being OK for home theater (this is because the surround effect gives ambience I presume). I think one could argue that movie sound track VLF tend to less discrete than music in general and are not as negatively impacted negatively by reverberation at low frequencies. Also, most of the VLF comes from the subwoofer, not from the surrounds, so wouldn't reverb in the low frequencies match better with the surround ambience in the higher frequencies? Total conjecture on my part...

Thanks,
Pete

[PeteD]

brucek wrote:

Perhaps best you start a new thread in the REW section to discuss your filters etc..... it will get more play there..

brucek

I am going there soon, I want to get some readings with less background noise, first. I was just playing around yesterday to generate the plots above.

Pete

[brucek]

I want to get some readings with less background noise, first

Use the multiple sweeps (up to 8) and longer sweeps (up to 1M) in the Measure screen to reduce noise considerably.

brucek

[JohnM]

PeteD wrote:

How do we use these waterfall plots, especially at low frequencies if the RT60 become meaningless?

The waterfall plots don't show RT60, they show the actual decay of each frequency across a band, which is what we need to see at low frequencies. RT60 is a single figure that summarises the decay across the entire band or some section of it (octave bands are fairly common), but a single number doesn't give any information about which specific frequencies are decaying slowly and which are decaying normally.

[DrWho]

Hey Pete, are those graphs showing the behavior after EQ?

[PeteD]

Hi Mike:

Yes, I quickly put together 5 or 6 filters, including +7 dB of boost at 20Hz, with a bandwidth of 120/60 - to compensate for cuts in the upper bass region, necessitated by a -12dB/octave x-over on my receiver and the fact that I have the same AE-IB -15s (4 in an IB, Fs of 32Hz, stiff spiders) that Chris Bee had before his recent upgrade. Without some boost, my receiver output is too low.

I will post some additional sweeps in a separate thread soon - my furnace is right behind my front wall/IB and it was on during measurements, along with some other stuff.

I am not sure of the relevancy of any readings from the RS meter below 10Hz, but using the curve (according to date stamp data) from the Shack which has factors down to 7Hz, I have output down to 2Hz (96 dB) on the curves shown above, with a dip at 5Hz.

Pete

[spudbudy]

hi gang just looking for a little advice on my recent setup of my BFD with the filters and a buttkickers with mains all in play.

[cporro]

ok, i'm new to this forum and hopefully this won't be a dissasterous post. i'd like to chime in on the idea of correcting a room via eq rather then acoustic treatments.

i know it's been a while but i started reading this post at the beginning.

my thinking goes like this. say you have a room mode at 60hz of +25db. you correct his with eq. now what do your ears hear?

if you are about 7 ft from your speakers you should first hear a direct signal. right? takes about 7ms to get to you since sounds travels about 1 ft/ms. so the first thing you hear is a signal with a major deficit at 60hz due to corrective eq.

after the initial direct signal all the other reflections come in. so for a mode you need the wave to go across the room and bounce back over itself. right? constructive interference (as i was taught). for 60hz we are talking about a room around 10ft across. so for the first standing wave to occur it should take longer then 7ms. maybe 20ms to hit a wall and come back.

ok, so you hear -25db at 60hz for the first 13ms. then the room mode kicks in and things are corrected.

seems like the same thing would happen on the way out as well. in other words the bass would be 13ms late arriving at its proper level and 13ms late decaying back down. think about this on something like a kick that is transient. i can tell you little things like 13ms on a kick are noticable.

13 ms might not seem like a long time but it sounds like one. just try putting 13ms delay on one channel and listen.

thats why i can't go for the the total eq solution. it's probably fine for minor stuff but not 20-30bd corrections. that just won't sound the same.

there are a few studio construction books (jeff cooper, f. altom everest) as well as bunch of recording pros that support this view.

[Wayne A. Pflughaupt]

Welcome to the Forum, Chris!

That's an interesting perspective, and it certainly seems logical. Perhaps our "salvation" is that I've never seen anyone with a 25-30 dB room mode, and we've seen hundreds of in-room measurements on this Forum. Have you been able to discern some delay when using an equalizer to tame a room mode?

Regards,
Wayne

[brucek]

Well, EQ is generally used where the room is dominated by the room's modal response. It certainly is only effective at the point of measurement (which would be the main listening postion). If a filter is applied that is the same Q and the opposite gain, then that resonance would be eliminated at that listening position. That's basically it.

brucek

[JohnM]

Hi Chris,

The direct sound is affected roughly as you describe, which is why EQ'ing the direct path is rightly frowned upon if not correcting for anomalies in the reproduction of the speaker itself, but you are forgetting the frequency you are considering. At low frequencies we are unable to perceive such short duration effects, our ears detect the total energy integrated over much longer periods than the time it takes sound to travel the length of a domestic room. The transient effects you detect in a kick drum, for example, are in the higher frequency content of the spectrum of that sound, not in the fundamental, and those higher frequencies which determine our perception of the transient are not being altered by the EQ filter. What is being corrected is the total energy which reaches us, which restores our perception of a balanced sound.

[cporro]

ok, maybe not a 30db mode. more like a +15 db at 60hz and a -15 db at 120hz. i guess it would be better to say unbalanced. that's what i've read and i can measure in my own room. sad.

as far as the low freq stuff goes. well, i don't know. i've moved a kick 20 ms and noticed a difference in the performance. i don't feel like my ear is doing an averaging at low frequencies.

if that were true (at low frequencies we can't easily detect transients) i wouldn't be able to here a truncated kick (7ms long) with a high pass killing everything above 100hz. i know i can. i just did it.

confusing mr. bigglesworth. and i just spent 1k on room acoustic treatment.

i feel like these arguments are coming out of 2 camps. one is theoretical and one is about having critical listening skills. i have a small foot in both i suppose.

[cporro]

and i must confess. i have never used eq to tame a mode (they are wild misundertsood beasts that must run free). maybe i should try before my acoustic treatments arrive. then post that.

[JohnM]

cporro wrote:

if that were true (at low frequencies we can't easily detect transients) i wouldn't be able to here a truncated kick (7ms long) with a high pass killing everything above 100hz. i know i can. i just did it.

I have no idea how you arrive at the conclusion that the energy integrating aspects of hearing at low frequencies should make you deaf.

cporro wrote:

i feel like these arguments are coming out of 2 camps. one is theoretical and one is about having critical listening skills. i have a small foot in both i suppose.

Chris, this is not being done as an academic exercise. If I didn't care passionately about good sound I wouldn't have spent the last several years writing analysis software to help people improve the sound they get from their systems and making the software available for free. I don't believe there is any conflict or incompatibility between striving for the best sound quality and striving to understand the acoustic principles and effects that contribute to it, quite the contrary.

cporro wrote:

i have never used eq to tame a mode (they are wild misundertsood beasts that must run free). maybe i should try before my acoustic treatments arrive. then post that.

Absolutely, we are great believers in measurement here.

[cporro]

not trying to give you a hard time. i'm one of those types that needs to reason things through or else i don't feel i know them. i start out with what i know (or think i know) and run through the scenario. if it doesn't add up i start asking questions like columbo.

btw, much much thanks for this useful program. i am just starting to get my head around it. right now i am loving the waterfalls. seems just about everything i need is in that one. dig how i can see the the issues with the speaker transform into issues with the room

[parsley]

Am I right in thinking that a waterfall plot helps to differentiate the response of the speakers (the first or early slices) from that of the room (the later slices)?

Mind you, for low frequencies e.g. 30 Hz where wavelength = 11.3 m and one cycle = 33 ms, is it possible to pick up the direct sound from the speaker and correctly identify its amplitude and frequency before the room response affects things at the measurement point?

Finally, in some of the waterfall plots shown previously (e.g. on page 2 of this thread) there seemed to be examples of the energy at a given frequency increasing a small amount with time. Is this likely to be a measurement/analysis artifact, or background noise, or non-linearity (e.g. something starts shaking and re-radiating energy at different frequencies)?

Thanks for the great software.

[brucek]

Am I right in thinking that a waterfall plot helps to differentiate the response of the speakers (the first or early slices) from that of the room (the later slices)?

No, the room is very much a part of the first slice (which is the same plot as the frequancy response) since the impulse response is gated (windowed). The next 29 slices are the decay of the signal over time in the room at the listening position. If you want the simple response of the speaker without the room influence, it's best to do a near-field away from boundaries (or go outside for the test).

brucek

[macrae11]

EDIT by Wayne Pflughaupt:

The next few posts (#97-#111) are some relevant posts that have been moved here from another thread that got sidetracked. They have been edited as needed to maintain relevancy to this discussion.



On another note, the love many people seem to have for DSP and room EQ is curious to me. Before I start, I must say that DSP and EQ can be a useful tool to setting up a system, but can be very misleading when used as a "be all end all" for deciding what sounds good.

First off, the goal, at least for me, and I would think the goal would be for most, is to recreate sound as close to what the original artist had intended to be heard. In this case the best scenario in a playback system is a completely transparent, uncoloured sound. Now I realize that not everyone has the means to design a system that is capable of this ideal, but the thought that this can be achieved with a mediocreor poor system and some EQ and DSP is severly flawed.

Now if your goal is not to recreate the most accurate representation of the source material, but to alter it to suit your aesthetic, such as having huge amounts of bass coming from your car system where you listen to hip hop, none of this concerns you.

Instead of writing out my argument myself I'm copying an explanation that is far better written that something I could compose, and states the issue much more clearly than I could. Note: I'm not referring to DSP used in crossover design, or creative desicions such as reverb. This is only reffering to the use of DSP and EQ to correct deficiencies in poor quality, audio equipment, speakers and rooms.

Here's the article. In particular I'm referring to section 3.
http://sound.westhost.com/articles/dsp.htm#s40


You cannot correct Time with Amplitude.

[tonyvdb]

macrae11 wrote:

On another note, the love many people seem to have for DSP and room EQ is curious to me. Before I start, I must say that DSP and EQ can be a useful tool to setting up a system, but can be very misleading when used as a "be all end all" for deciding what sounds good.

Now I realize that not everyone has the means to design a system that is capable of this ideal, but the thought that this can be achieved with a mediocreor poor system and some EQ and DSP is severly flawed.

You make some good points in your above post my only comment is on this section. You have to remember that an EQ or the so called room correction system is to get the sound coming out of the speakers in a room to sound like the original recording as close as possible. The problem is that room acoustics severally effects the sound, a "perfect room" should need little correction of the sound and a flat response is what you would get "if" the speakers are designed properly without any color. The issue is this is rarely the case. The other problem is that some recording studio's dont follow the rules and tweak there system so it is not flat to begin with and adds color to the sound even before it is mastered.