I find Holm invaluable foir this kind of thing. Because you can slide its window you can find out at what time the 410 Hz hole appears in the frequency response. This then tells you where to look in terms of distance away from the source. You can also window just reflections and find out what their frequency response is. Its an invaluable tool as far as I am concerned.
Holm seems more useful than what I've been using (more accurate and versatile) and I may pursue it. The learning curve is rocky but I'm part way there. It doesn't seem to lend itself to setting up subs, however.
Here is a spatially averaged plot of where I am at in the bass. Some EQ has been used here.
I have recently added a further 25 square feet of bass trapping to the existing 45 square feet (the prime of which is 10 inches of rockwool spaced 10 inches from the corner). The sound is more revealing than ever, and truly good, but not perfect.
Chasing good sound is like the proverbial drug. I'm not sure when or if I'll be satisfied. For now, I'll rest on the data. This plot is not perfect and I suspect it could be better in practice.
The EQ I have is inadequate but useful. It's discrete levels are too far apart as it seems I really am chasing +/- 1dB in some areas. I never thought a single dB either way was as important as I do now. It's just 'everything', when you know what I mean.
Here is a spatially averaged plot of where I am at in the bass. Some EQ has been used here.
I have recently added a further 25 square feet of bass trapping to the existing 45 square feet (the prime of which is 10 inches of rockwool spaced 10 inches from the corner). The sound is more revealing than ever, and truly good, but not perfect.
Chasing good sound is like the proverbial drug. I'm not sure when or if I'll be satisfied. For now, I'll rest on the data. This plot is not perfect and I suspect it could be better in practice.
The EQ I have is inadequate but useful. It's discrete levels are too far apart as it seems I really am chasing +/- 1dB in some areas. I never thought a single dB either way was as important as I do now. It's just 'everything', when you know what I mean.
Attachments
The first question you need to ask is always, "Do I need to optimize multiple seats or just one". Optimizing one seat is more straightforward.
I listen alone about 90 percent of the time, and no-one I know is as critical of the sound as I am, myself. So that is my priority. I have my mic on a rod on a tripod attached with a gimbal that I can rotate in 3 dimensions. When I spatially average, I make an 8 inch circle with the mic at about 1.5 rpm.
Do you have a picture of that measuring setup?
For such a small area I would run a Anti-Mode 8033 and be done with it.
For such a small area I would run a Anti-Mode 8033 and be done with it.
I have found this approach to be, by far, the most effective. To go for the best quantitative performance has always led me in the right direction. And on the contrary, I have heard systems design by ear and they virtually never cut it. Although, of course, the "ear" in question truely believes in its infallibility and omnipetent capabilities.
I've been relying on my ears for at least 30 years. For the more recent years of that time, I haven't been able to reach the goal relying on my ears alone. During the earlier years, I didn't even believe it was possible other than in my dreams...If only..., speakers werent bad then, only their usage 😉
I had to rely on my ears when working as a recording/mixing engineer. This taught me NOT to trust them.
When the systems were very bad, as they used to be, then listening was fine. It was coarse and unreliable, but with diligence you could get somewhere. At the level of details that we are looking at today this is no longer possible. The finer the detail and the more accurate the system gets, the less reliable listening will be.
Allen, if you are interested in the boundary issues you can also see my blog:
audio blog: A little more into boundary conditions
It also cantains a link to Markus's calculator. I've got some tables that are simpler to take in that the ones linked earlier. They are more like a "quick reference" guide.
I'm not so sure I'd kill the first contralateral reflections, but maybe. It sounds like your room is pretty small. A theory I have is that just killing the HF on the first contralateral reflection may fool your ear into thinking the room is actually larger than it is (and kill HF nastiness from diffraction). Maybe I need more coffee.......... More broadband might make it see huge or nearly outdoors but quiet. Envelopment it a good thing. I blame the lack of it for why I never seem to enjoy the sound at outdoor concerts. Small HF diffusion might be nice there as well--should make the room sound bigger while retaining the energy. Just some thoughts. I haven't tried all this out so I can't give you any first hand accounts. To me it seems like this should work. Broader band diffusion on the first ceiling reflection should be an ideal treatment because it's far enough away and well as a giant foam coffee table floor absorber to kill the floor bounce. At least in my head these make sense.
Dan
audio blog: A little more into boundary conditions
It also cantains a link to Markus's calculator. I've got some tables that are simpler to take in that the ones linked earlier. They are more like a "quick reference" guide.
I'm not so sure I'd kill the first contralateral reflections, but maybe. It sounds like your room is pretty small. A theory I have is that just killing the HF on the first contralateral reflection may fool your ear into thinking the room is actually larger than it is (and kill HF nastiness from diffraction). Maybe I need more coffee.......... More broadband might make it see huge or nearly outdoors but quiet. Envelopment it a good thing. I blame the lack of it for why I never seem to enjoy the sound at outdoor concerts. Small HF diffusion might be nice there as well--should make the room sound bigger while retaining the energy. Just some thoughts. I haven't tried all this out so I can't give you any first hand accounts. To me it seems like this should work. Broader band diffusion on the first ceiling reflection should be an ideal treatment because it's far enough away and well as a giant foam coffee table floor absorber to kill the floor bounce. At least in my head these make sense.
Dan
Last edited:
Well, Dan. I have a pair of 3' x 1.5' x 2" absorbers that I put over the contralateral points. Without them, one side is diffuse and the other is an in-wall fish tank which gives the highest amplitude reflective spike along the impulse response. It affects imaging somewhat.
My experience is that regardless of whether it is a direct reflection or which direction it comes from, killing the early stuff brings on 'the headphone effect'. I can't always (with my ears) tell which reflections I have just killed. From a psycho-acoustic point of view, I guess that the early reflections come as a bundled package.
By the headphone effect, of course, I don't mean that you can hear a pair of speakers, the image is fine. Horizontally and vertically. In fact, here's one for those who talk about vertical things...my waveguide group delay is sizeable and I cannot easily fix it in the crossover, and the drivers are separated on the baffle by a wavelength. I still seem to be able to tell how tall a singer is when I'm sitting, and as I stand, their head doesn't move from that spot.
Killing the earlies brings on clarity, but it could also be described as somewhat dead. I like the net effect though and I don't feel that incorporating later ambience compromises it. Consider the effect of electronic reverb as an example. When Dr Geddes points out that we want as little as possible up to a certain time after the direct, then make up for it with a nice reverberant field, I say yes, and more of it.
At the end of the day, I think this is what brings us closer to the acoustics of a larger room. If I imagine myself listening in a hall or other large room, less early reflections and more late ones seems likely. My listening room may not sound like a hall but the acoustics bear somewhat of a resemblance. From this I might venture the conclusion that when we say we like the sound of a large room, it is not the room itself that we like but the net effect to the sound. Clear but spacious.
My experience is that regardless of whether it is a direct reflection or which direction it comes from, killing the early stuff brings on 'the headphone effect'. I can't always (with my ears) tell which reflections I have just killed. From a psycho-acoustic point of view, I guess that the early reflections come as a bundled package.
By the headphone effect, of course, I don't mean that you can hear a pair of speakers, the image is fine. Horizontally and vertically. In fact, here's one for those who talk about vertical things...my waveguide group delay is sizeable and I cannot easily fix it in the crossover, and the drivers are separated on the baffle by a wavelength. I still seem to be able to tell how tall a singer is when I'm sitting, and as I stand, their head doesn't move from that spot.
Killing the earlies brings on clarity, but it could also be described as somewhat dead. I like the net effect though and I don't feel that incorporating later ambience compromises it. Consider the effect of electronic reverb as an example. When Dr Geddes points out that we want as little as possible up to a certain time after the direct, then make up for it with a nice reverberant field, I say yes, and more of it.
At the end of the day, I think this is what brings us closer to the acoustics of a larger room. If I imagine myself listening in a hall or other large room, less early reflections and more late ones seems likely. My listening room may not sound like a hall but the acoustics bear somewhat of a resemblance. From this I might venture the conclusion that when we say we like the sound of a large room, it is not the room itself that we like but the net effect to the sound. Clear but spacious.
Bummer about the fishtank, Allen. Mostly agree with your findings.
In a controlled setup, Contralateral reflections are almost always beneficial. If they need absorbed, it is most likely one of the other problems listed below, rather than being too early.
For readers who are confused by the term contralateral, it means Arriving from the side opposite the source. The first reflection from the left speaker to bounce from the Right wall (to the Right ear) is a contralateral reflection.
Dr. Geddes HT book (free chapters posted on his site) Gives a lot of good info about which reflections to keep, backed by real research.
The reasons they are generally desireable:
--Since they arrive at the opposite ear from the source, the brain easily de-correlates these reflections from the original signal
--They are usually lower in level than the first same-side wall reflection
--They are going to be better diffused than the first same-side reflection
--They are (typically) delayed enough to be useful for adding spaciousness
--Because of all the above, imaging is not harmed by strong contralateral reflections.
Same-side reflections often smear imaging, and facilitate comb-filter coloration.
Situations where contralateral reflections don't "work":
--When room is not fairly symmetrical in terms of both listener distance from sidewalls, and sidewall absorption properties
--When the reflectivity of wall surfaces is not full-spectrum - ie absorbing highs while reflecting midbass
--When rear wall is too close to listener, or if room is very narrow
If you are getting better results when absorbing the contralateral reflections, you probably have some of the room challenges listed above. Open floor plans, etc, real-world constraints prevent listeners from enjoying the benefits of those reflections.
It's also worth noting that (unmitigated) reflections arriving before the contralateral ones may mask them.
Once you experience a great reverberation tail on a recording (like pipe Organ in a cathedral) in a room with NO reflections within a 10+ ms window, and many reflections after that window, you'll be hooked. A two-channel setup can sound alarmingly realistic under these conditions.
Contrary to layman's logic, these reflections by themselves don't add reverberation on top of your recording, unless your room is like 40+ feet across, or is particularly reverberant in its own right. They only add a natural spaciousness that helps sounds to be more holographic and independent from the speakers, and creates a wider sweet spot that doesn't collapse every time you reach for your (insert beverage of choice here).
My 2 cents. -- Mark
In a controlled setup, Contralateral reflections are almost always beneficial. If they need absorbed, it is most likely one of the other problems listed below, rather than being too early.
For readers who are confused by the term contralateral, it means Arriving from the side opposite the source. The first reflection from the left speaker to bounce from the Right wall (to the Right ear) is a contralateral reflection.
Dr. Geddes HT book (free chapters posted on his site) Gives a lot of good info about which reflections to keep, backed by real research.
The reasons they are generally desireable:
--Since they arrive at the opposite ear from the source, the brain easily de-correlates these reflections from the original signal
--They are usually lower in level than the first same-side wall reflection
--They are going to be better diffused than the first same-side reflection
--They are (typically) delayed enough to be useful for adding spaciousness
--Because of all the above, imaging is not harmed by strong contralateral reflections.
Same-side reflections often smear imaging, and facilitate comb-filter coloration.
Situations where contralateral reflections don't "work":
--When room is not fairly symmetrical in terms of both listener distance from sidewalls, and sidewall absorption properties
--When the reflectivity of wall surfaces is not full-spectrum - ie absorbing highs while reflecting midbass
--When rear wall is too close to listener, or if room is very narrow
If you are getting better results when absorbing the contralateral reflections, you probably have some of the room challenges listed above. Open floor plans, etc, real-world constraints prevent listeners from enjoying the benefits of those reflections.
It's also worth noting that (unmitigated) reflections arriving before the contralateral ones may mask them.
Once you experience a great reverberation tail on a recording (like pipe Organ in a cathedral) in a room with NO reflections within a 10+ ms window, and many reflections after that window, you'll be hooked. A two-channel setup can sound alarmingly realistic under these conditions.
Contrary to layman's logic, these reflections by themselves don't add reverberation on top of your recording, unless your room is like 40+ feet across, or is particularly reverberant in its own right. They only add a natural spaciousness that helps sounds to be more holographic and independent from the speakers, and creates a wider sweet spot that doesn't collapse every time you reach for your (insert beverage of choice here).
My 2 cents. -- Mark
Last edited:
Sure is. I could attempt to replicate the other side using some lighter absorption over something diffusive. Another reason why it is tricky is that the loudest speaker axis is pointing across at that point. It is also 7ms beyond the direct so I'm still undecided. As I mentioned, the result of absorbing these reflections was difficult for me to differentiate from the others. I guess it could be due to the reflection time but diffusing it may just keep it in flight a little longer.
As I mentioned I modified my listening axis to best match the power response curve, and reconsidered my room treatments to match. I have noticed the stereo image has jumped out of the listening triangle where the material exists. I have heard it on two pieces so far. It's hard to describe the feeling when you hear a decisively struck low/middle register piano solo precisely two feet to the right of the woofer outside of the listening triangle. To say the least, I'm happy.
Assuming roughly an equilateral listening triangle, you must be pretty close--maybe 1.5 meters--to that sidewall in order for the arrival time difference to be that short. In any case, diffusion could work, as reflections from there will be reduced in level (compared to planar reflection). A small checkerboard pattern of absorbers will also provide a combo of diffusion+attenuation. If L-R differences in both color and time exist, you probably are better off absorbing it.
Ah-ha! Maybe disregard above ideas😱& leave well enough alone! What you are describing is indeed missing from most 'audiophile' systems. It's a feat that can't be achieved, regardless of components, unless the speakers-room-listener relationship is all working together.
The directivity of GedLee designs make that job easier, but it is still possible to achieve with other equipment & strategic room treatment.
-- Mark
Set up minimally relevant
Doesn't this condemn the seemingly endless tweaking, analysis, room treatments,praise of design, etc. to ambiguity at best, seeing as how you've confirmed that you "hear this on two pieces so far"? What about all the other music? Doesn't an honest interpretation of all this technical jargon and analysis/design require that all the music improve if indeed the acoustical signature of the room/speaker system is an accurate reflection of the requirements for accurate reproduction?
Seems to me that the constant banter about CD/polar response/power response/whatever at the end of the reproduction chain misses the mark, if different recording techniques in the SOURCE material can make such profound differences (which I've experienced myself) render end of chain playback tweaking and speaker designs apparently of minimal effect.
With my system, I've noticed dramatic differences between, for instance, music played back from cd/tape/vinyl, vs. the same music incidental on movie soundtracks, with the latter being enormously more enveloping and accurate to a live experience (on a competently engineered multichannel DVD), this with open baffle dipole mains and 7.1 surround, with distributed subwoofers.
John L.
Doesn't this condemn the seemingly endless tweaking, analysis, room treatments,praise of design, etc. to ambiguity at best, seeing as how you've confirmed that you "hear this on two pieces so far"? What about all the other music? Doesn't an honest interpretation of all this technical jargon and analysis/design require that all the music improve if indeed the acoustical signature of the room/speaker system is an accurate reflection of the requirements for accurate reproduction?
Seems to me that the constant banter about CD/polar response/power response/whatever at the end of the reproduction chain misses the mark, if different recording techniques in the SOURCE material can make such profound differences (which I've experienced myself) render end of chain playback tweaking and speaker designs apparently of minimal effect.
With my system, I've noticed dramatic differences between, for instance, music played back from cd/tape/vinyl, vs. the same music incidental on movie soundtracks, with the latter being enormously more enveloping and accurate to a live experience (on a competently engineered multichannel DVD), this with open baffle dipole mains and 7.1 surround, with distributed subwoofers.
John L.
I do not see this as true and in fact I would strongly disagree. If the recording is bad shouldn't a more accurate reproduction of it make it sound worse? If the reproduction somehow actually makes this bad recording sound great then something is really wrong, isn't it?
I just do not see your position as being true, which makes the rest of your post based on a faulty premise.
a bit reactionary don't you think?
Hi Earl...
Maybe I wasn't clear...
Not that it matters here but I think you've missed my point and my premise. If the goal is to provide a convincing reproduction of music as the artist intended, I think attention also should elucidate recording techniques that lead to improved soundstage, imaging, etc., all the issues deemed important and currently impossible to identify other than by trial and error.
perhaps some focus should be placed on addressing how to evaluate a recording (or other source) to determine if it is "good" or "bad" so that one doesn't waste money on lousy source material. How would I be able to tell this, unless the record label identifies these qualities in the release itself? This would then allow all the wonders of technical marvel to shine thru in all their glory, and we wouldn't have to listen to so many "bad" recordings sounding even worse thru super equipment.
cheers
Hi Earl...
Maybe I wasn't clear...
Not that it matters here but I think you've missed my point and my premise. If the goal is to provide a convincing reproduction of music as the artist intended, I think attention also should elucidate recording techniques that lead to improved soundstage, imaging, etc., all the issues deemed important and currently impossible to identify other than by trial and error.
perhaps some focus should be placed on addressing how to evaluate a recording (or other source) to determine if it is "good" or "bad" so that one doesn't waste money on lousy source material. How would I be able to tell this, unless the record label identifies these qualities in the release itself? This would then allow all the wonders of technical marvel to shine thru in all their glory, and we wouldn't have to listen to so many "bad" recordings sounding even worse thru super equipment.
cheers
Last edited:
- Status
- Not open for further replies.