Detailed Speaker Setup and Optimization

[Elliot G.]

I can understand skepticism here. This is partially why I opened this thread. I am being completly serious about the level of precision in setup that the human ear is capable of perceiving. I can hear 0.001 degrees difference in rake and certainly can hear it in azimuth. I don't have magic ears. You can hear it too. It is worth mentioning that if you just take a speaker and plunk it down and take 1 minute to get it roughly level such that it doesn't wobble then change the rake or azimuth by 0.001 or even 0.01 degrees you will likely not notice a change. It is when you get the speaker in a position that it is working with the room and it is getting time aligned that these differences of 0.001 degrees show up.

The interesting thing is that you don't have to have your head in a vice. As the speaker(s) become more and more coherent with the room, theirself and their mate the sound becomes quite open and free. you can sit many places and the music is great.

As scientist/engineer I used to think about audio in a scientific way. Things like: power cords can't possibly matter. Footers are just crazy talk. Speakers can be positioned by REW measurements. I no longer think this way. I use my ears as the instrument to tell me if something is correct or not. I don't know why our ears are sensitive to this level of precision. They just are.

@sbo6 asked if anyone had taken measurements during such fine positional changes. I have done some of this. REW will not get me there. At best it is a coarse tool to point me in a direction. Why? I think that REW measurements are really focused on giving us frequency domain information. The sweep is pretty long and the time window is also very open. Of course we do this because we want to capture everything the room has to say. However, our ears are picking up informaition in the low single digit microseconds. Our brain seems to be processing in the time domain. REW doen't have the ability to measure changes that are on this order of magnitude. It thinks in miliseconds.

Todd,
I agree I don't own a pair of speakers that only sounds good in one finite location and as I have "tuned" and adjsted further to my way to perfection LMAO , but rather have worked to get it is a precise as I can based on my patients level and the parameters of my real world room, the more enjoyable and engaging my system sounds. I have three listening charis in my room and all three seats sound great with the midlle of courst the prime local. I sit on oneside any time I have visitors and I can easily enjoyu and get in to the sound without feeling like I am only listening to a prt of the music. I think that may be reality with some products but it is certainly not all. I too have one of those levels and paid around mine. I like that i can use my cell phone to see what I am doing along with it. Very cool . I thank Stirling for that one and Kevin woldforheping use his at PAF two years ago.
THere is no cage match IMO it works and if you don't want to try it then don't.
One last thing and this Im sure will cause issues.When we set up speakers here ( this means me) locally its a three visit process and not a coupleof hours with tape measures and blue tape.
First visit- hook up and basic moving it around- leave let client put at least 100 hours of music on them
Second visit- serious set up and listening session to get the parameters correct and lock down settings.
Third visit- a few months later - to check the sound and if required make some minor adjestments as required.

As you and many others know there are no shortcuts and everyone can't or wont spend the time or the effort to go the whole distance.
WASP is a good beginning but IMO not a finished result. Let the shit storm begin

 

[Alrainbow]

Has anyone asked a speaker maker what other then there own brand do they like not so much love this is not cool. But like a bit even different types

 

[Elliot G.]

Has anyone asked a speaker maker what other then there own brand do they like not so much love this is not cool. But like a bit even different types

yes all the time. I am friends with a few speaker manufacturers and I am a big fan of their products.

 

[treitz3]

This is going to be another of those WBF Cage Match threads. I’m not going to get into the cage.
<snip>
And there is value in using a limited catalog for set up. Familiarity saves time.
<snip>

@Another Johnson - Good afternoon to you. I know that you are watching this from the side and my intention with this post is not to draw you back in but I would like to point out that you had mentioned twice in your post about "limited catalogs". Most of the folks I know who are in this hobby do not use a limited catalog of songs for setup. Maybe for initial speaker setup in a new room, just to find out where the bass/lower register response is the cleanest/most responsive but not for what these tools are used for.

I am assuming that the basics have already been covered and the general location of the speakers in a room have been established over time (as @Elliot G. alluded too), listening to a wide range of musical genre's. By this time, toe in is "usually" well established and general speaker/listening positions have also been established.

It is at this point, where these tools come in very handy to really dial in things further. Like @sbnx states, I do get the skepticism. I was never skeptical about precise setups, it's just that I always seemed to have other things that were more important to me than spending money on tools. Working on tone, improving this, improving that, cleaning up the noise, improving the streaming setup, trying different metallurgies, fuses, etc.. Over time, I have addressed many of the issues I felt were in need of improvement, so now I actually had no real desire to change anything else up......but simply improve the system as how I best saw fit.

This was the next step and honestly, I wish I had paid attention to it much sooner than I did.

After spending the time to do all of this last weekend to the finest degree I can (with the tools I currently have), I have had a blast just listening. No critiquing, no analyzing, no worries at all. Just listening to anything and everything. No limits and no limited catalogs of musical selections and yes, you can tell when you walk into the room. I don't really know how that works but you are definitely correct on that one.

As was mentioned earlier, this should not be a cage fight. @sbnx and I are simply sharing our observations. Folks are free to take this information how they will. We certainly are not forcing it upon anyone. We both just think that it is a very important part of the journey. It is definitely important to the end result as to what hits your ears. FWIW.

Tom

 

[sbnx]

I think @microstrip made some good points about measuring. I don't know every acoustic measuring device that is available. REW is free so most use it. Is it the best that can be done? I doubt it. It is called "Room EQ WIzard" which automatically implies it is primarily designed to help with frequency EQ curves and not for measuring speaker performance or to aid in finding optimal speaker locations.

I have purchased and used CLIO as well. It is primarily desinged as a tool to aid speaker designers. It has some interesting features. But again not really all that helpful in getting a pair of speaker dialed in.

One of the fundamental flaws with the above is that we are using one microphone in typically a center of the head location. Some time ago ( a few years) I was thinking about how one could mount multiple microphones in a fixed grid and then use this information to calculate timing differences. But I never built a prototype of this. I think this might work but would take quite a bit of effort. Great project for a graduate student.

Then recently we have the BACCH SSP system. This uses microphones that are actually placed in our ears to take the measurements. IMO, this is getting much closer. I have been wondering if during the speaker setup process if I used this to take measurements that I could pick up big differences in the interaural crosstalk. I think so. I think much of what is happening in a good speaker setup is that we are windowing out the room and minimizing the interaural crosstalk. But this is just a theory.

I am aware of some pretty sophisticated measuring systems that the big speaker companies use. They cost way to much money for an individual audiophile to buy (excluding billionaires). And one would still be left with the problem of how to interpret the measurements and which corresponding adjustment to make to the speaker.

We know what sounds righ and what sounds wrong. The human ear is capable of doing all of this. Just like our brain is capable of calculating the needed trajectory of a basketball and co-ordinating our muscles in order to hit a hoop not much bigger than the ball from quite a distance away. It does take some practice.

 

[sbnx]

I don't think I am standing alone in my thinking that we can hear in the low single digits of microsecond timing differences. Wilson Audio also clearly believes this as they are providing adjstments to the modules in the masterChronosonic and XVX that on this order of magnitude.

 

[microstrip]

(...) One of the fundamental flaws with the above is that we are using one microphone in typically a center of the head location. Some time ago ( a few years) I was thinking about how one could mount multiple microphones in a fixed grid and then use this information to calculate timing differences. But I never built a prototype of this. I think this might work but would take quite a bit of effort. Great project for a graduate student. (...)

In fact, when measured in a room, speakers should be measured multiple times with the microphone going through a grid - it is called a spatially averaged measurement. John Atkinson fully described his technique in a Stereophile article about speaker measurements.

 

[microstrip]

I don't think I am standing alone in my thinking that we can hear in the low single digits of microsecond timing differences. Wilson Audio also clearly believes this as they are providing adjstments to the modules in the masterChronosonic and XVX that on this order of magnitude.

As far as I remember David WIlson claims were around 5 microsecond.

But I must say it is not clear to me how you are using your digital level meter to carry the speaker adjustments - Wilson Audio supplies us with data that has been tabulated using pulse technique measurements.

 

[sbnx]

Yes. I think 5us is a good number for Wilson to shoot for in the adjustment mechanism of the speaker. Based on the research below pretty much all untrained listeners will be above this number. Some trained listeners will be below this number.

Here is a paper on time thresholds of human hearing that some may find interesting. They tested lots of variables that are in the paper if you want to peruse it. I have cut a few snips that cut to the chase.

The paper was published in 2019 in The Journal of the Acoustical Society of America. It is titled "Smallest perceivable interaural time differences". Google will return the paper.


"It is still noteworthy that thresholds for un-trained listeners are 2.6 times higher than for trained listeners."

"The average threshold ITD across nine trained listeners in a two-alternative forced choice task is 6.9??s"

"Two of the nine listeners performed significantly above chance at an ITD as small as 3.7??s."

"For 52 un-trained listeners, the average threshold ITD is 18.1??s"

 

[sbnx]

In fact, when measured in a room, speakers should be measured multiple times with the microphone going through a grid - it is called a spatially averaged measurement. John Atkinson fully described his technique in a Stereophile article about speaker measurements.

I am familiar with JA's grid technique where he takes several measruements in a 1m^2 grid. However, he is using this to average the frequency response. This is not what I was wanting to do.

I think 5 microphones would be enough. Once center, one left, one right, one up and one down. They must be precisely placed at exactly the same distance from the center. They must be fixed in position and all measurements taken at the same time.

The idea is to measure the time delta between the 5 microphones from the left speaker and then the right speaker. I think (key work is think as I don't actually know) that this would give sufficient information to tell the relative distances of the two speakers and also give information on which way to move the right speaker to align to the left. I think a 20kHz pulse would be best for this. But a higher frequency pulse would be better still.

I don't really know. These are just thoughts I had when I was thinking about how to get extreme acuracy of speaker position with measurements.

 

[sbnx]

I tried but I can't let these two recent thing go without writing about them. Both relate to speaker setup and how askew the "media" is on the subject. No wonder the average audiophile can't get it.

First is a well known youtuber. He has a large pair of Wilson speakers in his room. These have been there for a while and he even had the Wilson guy come set them up for him at one point. He raves about them constantly. Recently he got another "Reference level speaker" to play with/review in his room. Well, he moved the speakers in an placed them right in front of the Wilson speakers. This is a huge faux pas. No way are the new speakers going to sound their best in such a compromised position. Why does this matter? First, it is a free country and youtube is an open platform so he is free to post whatever he wants and say whatever he wants. But this is hugely irresponsible and demeaning to the brand that he put in front of the Wilson's. I hope his viewers are smart enough to realize how bad this is. I am sure he will give a big bag of excuses about how hard it would be for him to move the Wilson's out. But if I was the company representing the "New Reference Speaker" I would be seriously mad.

Ok, so the above is a pretty egregious example of non-detailed speaker setup by an amateur. Let's turn to the veteran mainstream press. Last night during halftime I thought "Hey, let's check out any new TAS videos they might have posted." So I found a video about the new YG Peak series with the subwoofer. I heard that speaker/subwoofer at the last show and thought it sounded pretty good in the hotel room so I was intrigued by what a reviewer would say. I want to thank TAS for the video review and thougth the video was well done and covered many of the aspects that a potential purchaser might want to know about. But at about 8:45 into the video the reviewer talks about how the YG rep come to visit and changed his setup. He says he moved the speaker about 1 foot diagonally and toed them out substantially from where he had them. He mentioned the speaker went from a ratio of about 0.8 to a ratio of 0.9. OK. OK. that is good to know. But then he says "Kind of impressive how much a small change in position can matter to the overall soundstaging". Wait! what!? Moving the speaker a foot diagonally and drastically altering the toe in is a HUGE change in position. How could he think this is a small change in position??

Sorry for the rant. I feel somewhat better but still concerned about what people are communicating regarding acceptable speaker setup practice.

 

[Al M.]

I tried but I can't let these two recent thing go without writing about them. Both relate to speaker setup and how askew the "media" is on the subject. No wonder the average audiophile can't get it.

First is a well known youtuber. He has a large pair of Wilson speakers in his room. These have been there for a while and he even had the Wilson guy come set them up for him at one point. He raves about them constantly. Recently he got another "Reference level speaker" to play with/review in his room. Well, he moved the speakers in an placed them right in front of the Wilson speakers. This is a huge faux pas. No way are the new speakers going to sound their best in such a compromised position. Why does this matter? First, it is a free country and youtube is an open platform so he is free to post whatever he wants and say whatever he wants. But this is hugely irresponsible and demeaning to the brand that he put in front of the Wilson's. I hope his viewers are smart enough to realize how bad this is. I am sure he will give a big bag of excuses about how hard it would be for him to move the Wilson's out. But if I was the company representing the "New Reference Speaker" I would be seriously mad.

Ok, so the above is a pretty egregious example of non-detailed speaker setup by an amateur. Let's turn to the veteran mainstream press. Last night during halftime I thought "Hey, let's check out any new TAS videos they might have posted." So I found a video about the new YG Peak series with the subwoofer. I heard that speaker/subwoofer at the last show and thought it sounded pretty good in the hotel room so I was intrigued by what a reviewer would say. I want to thank TAS for the video review and thougth the video was well done and covered many of the aspects that a potential purchaser might want to know about. But at about 8:45 into the video the reviewer talks about how the YG rep come to visit and changed his setup. He says he moved the speaker about 1 foot diagonally and toed them out substantially from where he had them. He mentioned the speaker went from a ratio of about 0.8 to a ratio of 0.9. OK. OK. that is good to know. But then he says "Kind of impressive how much a small change in position can matter to the overall soundstaging". Wait! what!? Moving the speaker a foot diagonally and drastically altering the toe in is a HUGE change in position. How could he think this is a small change in position??

Sorry for the rant. I feel somewhat better but still concerned about what people are communicating regarding acceptable speaker setup practice.

Great post. A foot? Fractions of an inch matter.

It seems unbelievable how much of speaker performance people throw away by not dialing in the speakers properly. And dialing in to perfection may take considerable time.

 

[Lee]

I don't think I am standing alone in my thinking that we can hear in the low single digits of microsecond timing differences. Wilson Audio also clearly believes this as they are providing adjstments to the modules in the masterChronosonic and XVX that on this order of magnitude.

It's actually been proven. David himself showed me research papers on the topic.

 

[Tangram]

It's actually been proven. David himself showed me research papers on the topic.

I also believe in careful speaker setup but I’m struggling with the idea that 1/16” difference is audible. Can you forward the names of the research papers David Wilson referenced? I may be able to track them down even if they are paywalled.

 

[Ron Resnick]

I’m struggling with the idea that 1/16” difference is audible.

I'll see your skepticism on 1/16", and raise you to 1/4".

 

[Lee]

I also believe in careful speaker setup but I’m struggling with the idea that 1/16” difference is audible. Can you forward the names of the research papers
David Wilson referenced? I may be able to track them down even if they are paywalled.

I don't have the paper citations. But you could look at Julian Dunn's papers on jitter audibility which show how excellent human hearing is on jitter for instance. That itself has been a moving target down to nanosecond and picosecond range.

 

[Lee]

I'll see your skepticism on 1/16", and raise you to 1/4".

Sorry, that's not right. Jim Smith and I found 1/4" differences clearly matter. I think we even got down to 1/8".

 

[Lee]

http://www.nanophon.com/audio/jitter92.pdf

 

[Ron Resnick]

Sorry, that's not right. Jim Smith and I found 1/4" differences clearly matter. I think we even got down to 1/8".

Interesting!

Nothing to be sorry about.

Maybe panel dipole speakers are different? I'm just not used to panel dipole speakers being that critical.

 

[Lee]

Interesting!

Nothing to be sorry about.

Maybe panel dipole speakers are different? I'm just not used to panel dipole speakers being that critical.

I don't think the type of speaker makes setup any less critical. I have a long history with Magnepans and they need precise placement.