Room Curves, System voicing, and related thoughts

[CharlieLaub]

It may not be obvious, but this does fit more or less with my revised target concept, based on Linkwitz's studies with Don Barringer and his Version 2 curve.

Hi Jon,

I thought that the Linkwitz/Barringer study was done for the Orion or similar dipole speaker. The power response is bound to be very different for that kind of source, versus your speaker. A monopole would likely have a sagging power response in the upper octaves without any reduction in input voltage like what is shown above.

Given that the perceived total sound at the listening position comes from both the direct and indirect (room reflected) sound, do you think that their "best sounding" curve would be "best" for this speaker?

-Charlie

[JonMarsh]

Hi Jon,

I thought that the Linkwitz/Barringer study was done for the Orion or similar dipole speaker. The power response is bound to be very different for that kind of source, versus your speaker. A monopole would likely have a sagging power response in the upper octaves without any reduction in input voltage like what is shown above.

Given that the perceived total sound at the listening position comes from both the direct and indirect (room reflected) sound, do you think that their "best sounding" curve would be "best" for this speaker?

-Charlie

A very valid question, Charlie. For a monopole system, due to the cabinet design and drivers, there's a pretty wide dispersion. Monopoles will be inherently more room sensitive. One of slightly lesser roll off curves may be more suitable- but frankly, I plan to add a toggle switch and be able to have at least two settings.

Could be the version 1 curve is best for these. Withe my experience with the Ardent and Modula Xtreme, I don't think dead flat is best. The wider the dispersion, the less I think so!

And I'm not going to finalize the voicing until they're over at GF's, and I have the new DAC. But they sound very good at home like this with a wide range of material, and are quite comfortable at higher volumes. Y'all can pick the voicing to pieces at the Northern CA DIY, if they get there! With external crossovers, it's easy to fix!

[Paul W]

Jon,

Regarding the "listening curve" here's a very interesting thread.

http://www.hometheatershack.com/forums/rew-forum/64094-psychoacoustically-correct-room-correction.html#ixzz2MKVo8oBN

Bob Katz is probably worth reading!

[JonMarsh]

Yes, I've read some of Bob's stuff in the past, and I do have REW, what I'm still lacking is some more spare time to work with it. I suspect I'll find a lot to like, based on the feature set update for REW 5. OTOH, I've had so much to get done lately, my focus in the near term has been sticking with tools that I've somewhat mastered... in principle REW should be able to do a lot more from what I've read than Fuzzemeasure, but the latter has the stuff I need to measure drivers, evaluate baffle radiation and so forth, and get some speakers designed!

Also, when playing with some DSP EQ in my LIO-8, my experience (which mirrors some other folks on Hometheater Shack) is that I tend to prefer a straight line curve with a slight downtilt over one like the above posted (which at -6 dB at 20 kHz is probably fine for HT but not music- and a lot of the emphasis for REW is HT)- though I'm sure how you measure that, the room acoustic and RT will all influence perception.

Lots of things to ponder when there's more time-

[Paul W]

I've been listening to a straight tilt for a couple of years but, though a 3-5db straight tilt sounded better than flat, it didn't reduce sibilance to a satisfactory level. So, I tried a slight variation of Bob's curve by boosting 1k to almost flat...about 1db straight tilt from 100-1k folllowed by a slightly steeper straight tilt from 1k to my original starting point at 10k. Introducing that small knee reduced apparent sibilance even though that range is well above 1k. Guess it makes sense since levels are all relative.

[JonMarsh]

Well, the important thing here, is that it sounds like you've found something that works for you. And I'll keep in it mind when looking at this in the future- thanks for sharing your experience. As much as source material varies in it's tonality and likely mastering, probably any one curve is going to be a bit of a compromise, which is another reason I expect to probably have a level/curve switch of some type in the final crossover, probably labeled something like "Audiophile/Standard". :W

[gbegland]

I was at Bob's a few weekends ago doing some rewiring and other items. I listened to his new digitally processed monitor set-up referenced in that thread. I thought they sounded a little "polite" for my tastes. I would have lifted the top end to be more around -3 to -4 dB rather than the -6 he's running now. He uses Revel Gem2 paired with JL subs. They used to be crossed with a custom Marchand module and EQ'd with a special Meyer Sound parametric model. Now it's all digital "in the box" run from his mastering computer. He sits pretty close to the monitors are they are well away from the walls, so power response is slightly less of an issue there. Really like large headphones in a sense.

It sounds very nice, but honestly those prototype monitors we just built with the OEM Raal and the Neo8-S are far more resolving....sssshhhhh, don't tell Bob.

I think my house curve is a little more like Pauls, with a downward slope right from the bass range, not up at 1KHz, but the whole slope only ends around -3dB.

Greg

[5th element]

Sibilance typically occurs at about 5k, but depending on the word uttered can cover from around 2-10k. This helps explain why mid/bass breakup can influence the sound significantly if handled incorrectly and people usually assume that it's the tweeter.

For what it's worth, with my OBs I ended up with a downward tilt, but with the monopoles + CD tweeter with 6.5" waveguide, flat sounds fantastic without sounding anything close to too bright, or having trouble with sibilance. In fact the pleasing way in which the WG handles sibilance is one of its strong points. I must add though, that getting the tweeter level right with the WG seemed far more critical than with other designs otherwise the sound was way too hot and in an unpleasant in your face kind of way, whereas with non WG designs things just sound a little bright.

[Paul W]

Yeah, I didn't mean to imply "one size fits all". Different DI in different rooms will likely require different curves...and recordings really are all over the map. Even though it makes some sense, I didn't expect to hear such a difference in relative sibilance. Overall, early results sound very relaxed and natural, so I'll be tweaking the approach for a while.

[JonMarsh]

Sibilance typically occurs at about 5k, but depending on the word uttered can cover from around 2-10k. This helps explain why mid/bass breakup can influence the sound significantly if handled incorrectly and people usually assume that it's the tweeter.

For what it's worth, with my OBs I ended up with a downward tilt, but with the monopoles + CD tweeter with 6.5" waveguide, flat sounds fantastic without sounding anything close to too bright, or having trouble with sibilance. In fact the pleasing way in which the WG handles sibilance is one of its strong points. I must add though, that getting the tweeter level right with the WG seemed far more critical than with other designs otherwise the sound was way too hot and in an unpleasant in your face kind of way, whereas with non WG designs things just sound a little bright.

I think you've personally identified an important parameter, especially if not listening in the near field, that the power response and dispersion are quite important to the perception as well as the nominal axial response.

I also find that how clean the rest of chain behaves is quite important, too- for example, between one popular $1,000 DAC and the NAD M51, I find there's a substantial difference in what I call resolution or grunge (or lack of it!) in the presence region to high end, which mitigates to degree the usual effects of brightness. In other words, what might be perceived as a little bright or forward is MUCH more annoying if that is bringing distortion products to the fore in a frequency range the ear is quite sensitive, but is much less objectionable (to me) if everything is relatively crystal clear/pristine.

The improved mastering in some recent CD issues of really old records seems to highlight this- (listening to some old Hollies and Steeleye Span re-issues) yes, the tonal balance and mix and presentation is still that of old recording gear from the late 60's or early 70's, but now, it's more as if I'm hearing the analog master tape than a CD dub of it. Not all the way there, by any means, but often much closer. And as a lover of many kinds of old as well as new music, that's a big deal to me, and why I'm shelling out for an even more expensive DAC and possibly one of their reclockers (I get a special deal on it equivalent to the combined purchase for the rest of 2013 if I decide to "pull the trigger").

This is a complex topic, and while there's a lot of people trying to find a neat formula for how to approach it, I think it's rather dependent on the room acoustics, treatments, program material, power response, etc- but I'm enjoying the fact that a number of you are thinking in these terms and working out your own solutions- this is something to discuss more, perhaps consolidate some of these posts in a separate thread and continue on the topic as time and energy permit-

Thanks Greg, Fifth Element and Paul!

[jim1961]

Jon,

Regarding the "listening curve" here's a very interesting thread.

http://www.hometheatershack.com/forums/rew-forum/64094-psychoacoustically-correct-room-correction.html#ixzz2MKVo8oBN

Bob Katz is probably worth reading!

The black line is the original. I added the red to illustrate about what my personal preferences are. Of course this varies somewhat based on the type of music im listening to. For me, -6db @10k is a bit much. And I like a bit more (1-2db) in the 20-35hz range than I am indicating here (I really like to FEEL the music)

[jim1961]

I think you've personally identified an important parameter, especially if not listening in the near field, that the power response and dispersion are quite important to the perception as well as the nominal axial response.

I also find that how clean the rest of chain behaves is quite important, too- for example, between one popular $1,000 DAC and the NAD M51, I find there's a substantial difference in what I call resolution or grunge (or lack of it!) in the presence region to high end, which mitigates to degree the usual effects of brightness. In other words, what might be perceived as a little bright or forward is MUCH more annoying if that is bringing distortion products to the fore in a frequency range the ear is quite sensitive, but is much less objectionable (to me) if everything is relatively crystal clear/pristine.

The improved mastering in some recent CD issues of really old records seems to highlight this- (listening to some old Hollies and Steeleye Span re-issues) yes, the tonal balance and mix and presentation is still that of old recording gear from the late 60's or early 70's, but now, it's more as if I'm hearing the analog master tape than a CD dub of it. Not all the way there, by any means, but often much closer. And as a lover of many kinds of old as well as new music, that's a big deal to me, and why I'm shelling out for an even more expensive DAC and possibly one of their reclockers (I get a special deal on it equivalent to the combined purchase for the rest of 2013 if I decide to "pull the trigger").

This is a complex topic, and while there's a lot of people trying to find a neat formula for how to approach it, I think it's rather dependent on the room acoustics, treatments, program material, power response, etc- but I'm enjoying the fact that a number of you are thinking in these terms and working out your own solutions- this is something to discuss more, perhaps consolidate some of these posts in a separate thread and continue on the topic as time and energy permit-

Thanks Greg, Fifth Element and Paul!

Many good points here, especially the ones concerning the room response playing a noticeable factor. In my mind, buying better gear should only be considered when the room is treated properly. What many people fail to realize (perhaps not people here) is that much of what they hear isn't the speaker (direct response), but reflections both separate from the direct sound and their inherent interaction and augmentation (in many cases malevolent) with the direct response itself.

Let me illustrate a phenomena I experienced. I was hearing something in the 2k-6k range that was really bothering me. At certain times, a note would hit in that range of frequencies and make me want to ring my ear. They sounded louder than they should, but my testing software said those frequencies were in line. So I then got caught in those charts that describe the human hearing curve, and thought maybe I was just ultra sensitive to those frequencies. At the time, I had an Parametric EQ in the signal path so I adjusted down those frequencies. Ah, yes, that was better. But after some time of additional listening, it became apparent I hadn't really addressed the problem at all, just made it less audible. In the end, I figured out that I was hearing comb filtering and further room treatment actually addressed the problem. I then turned up those frequencies I had previously diminished with the EQ and now they sounded fine.

I am sure folks here are aware of the "loudness wars" that permeate the CD industry. These Cd's are going to have a treble range that would lead one to believe that a stronger high end roll off would be quite welcome. On the other hand, vinyl lovers, especially those with tube gear have become accustomed to a warm and natural high end roll off. So in my experience anyway, there is a wide diversity of high end characteristics in source material, and when you try to get all your source material to sound a particular way, its problematic.

Now there have been /are some very nice CD remasters of older material that don't suffer from "loudness war syndrome" that have a more analog quality about them. I agree with you Jon on this.

At the end of the day, I don't see how one formula or approach can make everything sound right. But if you all are like me, you take your best sounding material and optimize your system, speakers and room to make that material sound its best. What else can we do? This has the unfortunate consequence of me favoring inferior music with great sound over better music with inferior sound (inferior masterings)

Yes, it is a complex issue. And the more I learn, the better I seem to be able to sort it out and know how to address specific issues. Threads like this serve this goal well

[JonMarsh]

Room Curves, System voicing, and related thoughts

There's been some interesting things shared OT in other threads that I'd like to be sure don't get lost and have a chance for exposure to a wider audience- so I'm starting this thread by copying/moving some of those posts into this thread, where we'll have a more visible locale for discussion in at the tag end of a 25 or 43 page project thread!

Obviously, we welcome discussion and experiences from all interested parties- how have you dealt with your room setup issues, speaker voicing based on system and environmental considerations, etc. I think this is topic we don't touch on often enough, but which at a certain stage of development become critical to moving the goal posts forward and taking things in both HT and music playback to the next level.

Welcome...

[CharlieLaub]

One thing that I have always been curious about, but I don't find is discussed along with the loudspeaker response curve data, is the contribution from the room. I can imagine the situation in which the room is quite live, making a flat system sound too bright and forward and fatiguing. Similarly, someone could have a room full of heavy carpet, overstuffed furniture, etc. and flat might sound neutral or even a little dead.

So, with that said, can there be a "best" frequency response if the reverberant field is not also considered?

-Charlie

OK I am reading through some of the linked posts at HomeTheaterShack and they seem to be touching on this subject. There is a lot of material that I need to take in on this for sure!

[Face]

With 10 bass traps, a few diffusers, area rug, and potted plants around, a curve like that would sound dull in my room.

[jim1961]

One thing that I have always been curious about, but I don't find is discussed along with the loudspeaker response curve data, is the contribution from the room. I can imagine the situation in which the room is quite live, making a flat system sound too bright and forward and fatiguing. Similarly, someone could have a room full of heavy carpet, overstuffed furniture, etc. and flat might sound neutral or even a little dead.

So, with that said, can there be a "best" frequency response if the reverberant field is not also considered?

-Charlie

OK I am reading through some of the linked posts at HomeTheaterShack and they seem to be touching on this subject. There is a lot of material that I need to take in on this for sure!

In a sense, what the loudspeaker is doing is irrelevant considering what you hear at your listening position is what is heard, not some 5ms gated response 1m from the speaker. That said, certainly making the speakers response as flat as possible is a desirable design goal. But what is trickier than a flat speaker is a flat responding room. This is augmented by the power response of the speaker which usually wont be as flat as its on-axis response. So in this sense, for the response of the room to sound flat, the room response would have to be an inversion of the power response and not flat as one may think. Throw in SBIR and modal data and you have a complex situation indeed.

[CharlieLaub]

In a sense, what the loudspeaker is doing is irrelevant considering what you hear at your listening position is what is heard, not some 5ms gated response 1m from the speaker. That said, certainly making the speakers response as flat as possible is a desirable design goal. But what is trickier than a flat speaker is a flat responding room. This is augmented by the power response of the speaker which usually wont be as flat as its on-axis response. So in this sense, for the response of the room to sound flat, the room response would have to be an inversion of the power response and not flat as one may think. Throw in SBIR and modal data and you have a complex situation indeed.

Yes, I agree with everything you say, in principle.

I've more or less figured out the crossover design part of the loudspeaker design endeavor, measuring on axis response, etc. Next I would like to be able to calculate the directivity and power response, I guess from multiple off axis measurements or from a diffraction model plus the crossover functions. Can you suggest good technical reading on that subject?

-Charlie

[jim1961]

Yes, I agree with everything you say, in principle.

I've more or less figured out the crossover design part of the loudspeaker design endeavor, measuring on axis response, etc. Next I would like to be able to calculate the directivity and power response, I guess from multiple off axis measurements or from a diffraction model plus the crossover functions. Can you suggest good technical reading on that subject?

-Charlie

http://www.htguide.com/forum/showthr...n-Study/page25 post 844

Jon just posted something there having to do with crossover design with off axis in mind.

I am more a room designer than x-over designer, so I am afraid I cant recommend anything to you. Hopefully, someone well versed will come along and do so.

Edit: Re-reading your post it triggered an additional thought. Since the reason we are concerned with directivity and power response in the first place is due to the ultimate consequences of how those delayed signal paths and their spectral content affect the listening experience, it seems to me that the two (speaker and room) really must be considered together. For example, if the power response was really tilted towards high frequency dispersion but the room had the character of absorbing high frequency content more than the rest, it may coalesce into a neutral end result.

[5th element]

Let me illustrate a phenomena I experienced. I was hearing something in the 2k-6k range that was really bothering me. At certain times, a note would hit in that range of frequencies and make me want to ring my ear. They sounded louder than they should, but my testing software said those frequencies were in line. So I then got caught in those charts that describe the human hearing curve, and thought maybe I was just ultra sensitive to those frequencies. At the time, I had an Parametric EQ in the signal path so I adjusted down those frequencies. Ah, yes, that was better. But after some time of additional listening, it became apparent I hadn't really addressed the problem at all, just made it less audible. In the end, I figured out that I was hearing comb filtering and further room treatment actually addressed the problem. I then turned up those frequencies I had previously diminished with the EQ and now they sounded fine.

I've experienced a similar thing though for different reasons. I developed a case of hyperacusis in my left ear and this was only noticeable, at first, when listening to music.

When this first came on it was very mild and I attributed it to my loudspeakers as I was always making changes to them.

Similar to you, I played around with parametric EQ around the presence region for the ears and this helped somewhat, but didn't stop it completely.

As the symptoms got worse I then realised that it was my ear and not the loudspeakers and I noticed that listening over quality headphones was a lot more tolerable than with loudspeakers.

Since then I've played around with various waveguides/tweeters and systems and I've found that having a flat, smooth, well controlled power response is very helpful in keeping the acusis under control. Whenever I swap to a loudspeaker that has a classic wide narrow wide pattern the pain comes on. I always jump back to my own system just to check that I'm not imaginging it and boom, ahhh, my ears are happy listening to this all day, but the other systems, that are very flat on axis, irritate me like no other.

Now having a standard format dome tweeter, but with a small mid/bass and crossed low, so that you pretty much do away with the wide narrow wide deal, also helps, but not as much as going to a CD waveguide.

I haven't played with any types of room treatment, but it is very possible that it could very well have a big impact on this too, as it's most certainly not anything in the direct, on axis, sound that is triggering the acusis, but something in the off axis.

This of course is maybe a somewhat specific issue, but lots of people suffer from one form of hyperacusis or another, such as tinnitus related hearing sensitivity in specific frequency bands and the forever laboured 'listening fatigue' effect is something people talk about all the time. Now I wouldn't mind betting that some of the systems that would have me running from the room would be loved by those who are after a very forward presentation, but for a relaxing, fatigue free, listen all day, type of sound, the over all off axis response, to a certain degree, might even be more important than the on axis.

Naturally one doesn't want to deviate too far away from flat in the direct sound, but I'd happily trade the odd dB of flatness axially, for a nice power response if possible. And if it's not possible then I'd change the loudspeakers design so that it is, that's if you want me to listen to it.

[CharlieLaub]

http://www.htguide.com/forum/showthr...n-Study/page25 post 844

Jon just posted something there having to do with crossover design with off axis in mind.

My opening post was actually a reply in that thread, but Jon broke it out to form this separate thread with a more focused subject. It seems to be that in post 844 he's not addressing power response or room response per se, but is using the 30 degrees off axis position as the "listening axis", e.g. the speakers will be somewhat far apart and positioned straight ahead (not toed in) so that where the listener is seated is then "off axis" compared to where the drivers are pointing. I think that is a very good idea, and I have done something similar with a 2-way open baffle speaker that I have been working on (and will likely bring to the NorCal DIY event in April). I have attached a pic of the frequency response family below.

I am more a room designer than x-over designer, so I am afraid I cant recommend anything to you. Hopefully, someone well versed will come along and do so.

Edit: Re-reading your post it triggered an additional thought. Since the reason we are concerned with directivity and power response in the first place is due to the ultimate consequences of how those delayed signal paths and their spectral content affect the listening experience, it seems to me that the two (speaker and room) really must be considered together. For example, if the power response was really tilted towards high frequency dispersion but the room had the character of absorbing high frequency content more than the rest, it may coalesce into a neutral end result.

This is exactly the point that I had in mind - what you hear is the combination of the loudspeaker and the room. What you hear from the room is a product of the loudspeaker's off axis response and the room's frequency dependent adsorption/reflection properties. The size of the room also plays a role, as larger rooms will have less relative power compared to the loudspeaker's direct response than will smaller rooms.

This all seems to indicate that it is very difficult to make a good sounding loudspeaker without knowing something about the space in which the loudspeaker will reproduce the input signal. There could be some kind of switchable "tilt" control that might allow the speaker to be better matched to the room, and this is (probably) most easily implemented in an active system.

-Charlie

Below is a screen shot of the off axis responses for my MWOB loudspeaker system. Note that the listening position will be at about 25-30 degrees off axis, and the response is flat or slightly down tilted for that location. Closer to on axis the response rises above 1kHz and farther off axis the response tails off. Hopefully this will result in a balanced power response.

[jim1961]

Here is an interesting way to look at an ETC. Some call it "slicing". Typically, the ETC is spectrally blind, that is it gives you time and magnitude domains, but no spectral info. But by limiting the frequencies used by the ETC, you can plot what different ranges are doing as I have done in the above graph.

Blue = 4.2k
Green = 2.6k
Yellow = 1.6k
Orange = 1k
Red = 668hz

These are 200us smoothed 1 octave in bands.

Just another tool, but one I seldom see other people use.

[Jonasz]

I've experienced a similar thing though for different reasons. I developed a case of hyperacusis in my left ear and this was only noticeable, at first, when listening to music.

When this first came on it was very mild and I attributed it to my loudspeakers as I was always making changes to them.

Similar to you, I played around with parametric EQ around the presence region for the ears and this helped somewhat, but didn't stop it completely.

As the symptoms got worse I then realised that it was my ear and not the loudspeakers and I noticed that listening over quality headphones was a lot more tolerable than with loudspeakers.

Since then I've played around with various waveguides/tweeters and systems and I've found that having a flat, smooth, well controlled power response is very helpful in keeping the acusis under control. Whenever I swap to a loudspeaker that has a classic wide narrow wide pattern the pain comes on. I always jump back to my own system just to check that I'm not imaginging it and boom, ahhh, my ears are happy listening to this all day, but the other systems, that are very flat on axis, irritate me like no other.

Now having a standard format dome tweeter, but with a small mid/bass and crossed low, so that you pretty much do away with the wide narrow wide deal, also helps, but not as much as going to a CD waveguide.

I haven't played with any types of room treatment, but it is very possible that it could very well have a big impact on this too, as it's most certainly not anything in the direct, on axis, sound that is triggering the acusis, but something in the off axis.

This of course is maybe a somewhat specific issue, but lots of people suffer from one form of hyperacusis or another, such as tinnitus related hearing sensitivity in specific frequency bands and the forever laboured 'listening fatigue' effect is something people talk about all the time. Now I wouldn't mind betting that some of the systems that would have me running from the room would be loved by those who are after a very forward presentation, but for a relaxing, fatigue free, listen all day, type of sound, the over all off axis response, to a certain degree, might even be more important than the on axis.

Naturally one doesn't want to deviate too far away from flat in the direct sound, but I'd happily trade the odd dB of flatness axially, for a nice power response if possible. And if it's not possible then I'd change the loudspeakers design so that it is, that's if you want me to listen to it.

I have the same problem as you and I totally agree with your theory about controlled directivity. That's probably why I'm so happy with the NaO Note.

[jim1961]

This is exactly the point that I had in mind - what you hear is the combination of the loudspeaker and the room. What you hear from the room is a product of the loudspeaker's off axis response and the room's frequency dependent adsorption/reflection properties. The size of the room also plays a role, as larger rooms will have less relative power compared to the loudspeaker's direct response than will smaller rooms.

This all seems to indicate that it is very difficult to make a good sounding loudspeaker without knowing something about the space in which the loudspeaker will reproduce the input signal. There could be some kind of switchable "tilt" control that might allow the speaker to be better matched to the room, and this is (probably) most easily implemented in an active system.

-Charlie

Lets take an example and see where this goes. The power response is relevant because one supposes the off axis response bounces off a reflective surface and reaches the ear at some later time. And that the spectra of the reflection(s) propagated by the power response will be true in their spectra after they reflect. Lets also suppose we are really only talking about frequencies above 300hz or so given baffle step phenomena.

The above assumes some things that are not necessarily true. In a treated room for instance, 1st reflection points are usually either absorbed or redirected. In either case, the power response isnt going to be linearly represented spectrally as a reflection arriving at the listening position. In one case (redirected) it doesnt arrive at all as a early high gain reflection, and in the second case (absorbed), assuming your absorbing in a broadband manner, the arriving reflection has little energy and thus little effect.

So it seems to me that for the most part, power response linearity is important in untreated rooms much more than treated ones.

But even in the case where the off axis response bounces off a bare wall, your not necessarily going to get a linear spectrum bounce from it. The ETC I presented a few posts ago illustrates clearly that when you bounce sound around, you dont get a mirror spectrum bounce from it. Surfaces all around the room depending on reflectivity, size and the delay change the spectrum content considerably.

So, like a few posts ago when I said that a flat on axis response is a good design criteria, I would still say that about the power response as well. But clearly the power response as it pertains to how it reflects around the room puts it through a cheese grinder so to speak and by the time you hear it, it isnt the same at all as it measured at the speaker.

[Paul W]

Jim,
Very interesting ETC. Looks like early reflections are very well controlled yet there is something big at 24ms??? What software are you using?

[jim1961]

Jim,
Very interesting ETC. Looks like early reflections are very well controlled yet there is something big at 24ms??? What software are you using?

I am using a Dayton Omnimic. The 24ms peak is there by design. I am emulating a LEDE / RFZ model whereby you have an ISD gap followed by a termination peak. The modal requires / suggests at least a 20ms gap @ -20db or better followed by a termination peak of not less than -12db.

My normal ETC looks like this:



Looking at this as it relates to the thread topic, how much does the power response matter if its reflections are > -20db or -25db down?