• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

The precedence effekt - very important för the sound enthusiast

Tangband

Major Contributor
Joined
Sep 3, 2019
Messages
2,994
Likes
2,789
Location
Sweden
This is something thats very important to learn about, if you are serious about your hifi.

The ear and your brain is funtioning very different compared to how a microphone works.
The brain selects the sound, the microphone dont.

https://en.m.wikipedia.org/wiki/Precedence_effect

Read very carefully about the precedence-effect and then draw conclusions about where to place your loudspeakers in a normal room.
The two loudspeakers should be placed so that the destructive sound bounce from the wall behind the loudspeake is MORE than 2 ms from the direct sound coming from the speakers. 1 ms is 34,3 cm traveling distance for sound. 2 ms is the double, about 68 cm.

The conclusion of this is for a good sound, where you can discover a good natural spatial experience, is that your loudspeakers should be placed more than 34 cm away from the wall. In my experience , the sound benefits even more with a longer distance, you can begin trying about 43 cm or more from the loudspeakerbaffle to the back-wall.
 

mixsit

Member
Forum Donor
Joined
Jul 20, 2019
Messages
75
Likes
23
The two loudspeakers should be placed so that the destructive sound bounce from the wall behind the loudspeake is MORE than 2 ms from the direct sound coming from the speakers. 1 ms is 34,3 cm traveling distance for sound. 2 ms is the double, about 68 cm.
The conclusion of this is for a good sound, where you can discover a good natural spatial experience, is that your loudspeakers should be placed more than 34 cm away from the wall. In my experience , the sound benefits even more with a longer distance, you can begin trying about 43 cm or more from the loudspeakerbaffle to the back-wall.
I'd like to offer a counter to this. The the part of the precedence effect you're referring to is the problem of the delayed version of a source being perceived as being part of the source.
While the combined effects of direct plus delayed paths begin, well anywhere greater than zero MS, they continue on into the 20 - 30MS range (-transient content earlier, sustained taking longer) before sounding like delayed second source.
So we have this stuff bouncing of all the walls, ceiling and the behind speaker wall -that basically blurs the source in time, that can't be 'fixed by speaker placement. Thankfully, the another part of the precedence -source location via first arrival, (presumably/depending?) remains intact.
The typical solution is treatment to reduce these first reflections.
Which puts speaker placement back to it's interaction in the lower bands -where few but studios attempt to address, for destructive effects on frequency response.

How (why?) did I become such a nerd' like this :>) ..Boning up 40 years in music and home recording figuring I best lean my tools :>)
Thanks
 

mikewxyz

Member
Joined
Jun 19, 2019
Messages
37
Likes
39
This topic (also called the Hass effect) is covered extensively on the Gearslutz site. Here is one example -
https://www.gearslutz.com/board/stu...nderstanding-precedence-effect-etc-shift.html

You can search for more threads. There are some competent people who frequent that forum. My take away from reading several of these threads is that 1) you need to know what model you are trying to build for your acoustic space (LEDE, RFZ, etc). 2) Recording rooms are different and have different requirements than listening rooms.
 

JanRSmit

Member
Joined
Jan 21, 2019
Messages
54
Likes
21
Well the 2msec is in fact true. When applied to monopole loudspeaker placement, it indeed implies to place the speaker baffle at least 1m before the wall behind the speaker (in english often referred to as the front wall).
Easily tested.
This is true for the part of the spectrum radiating also around the speaker. The lower part of the spectrum. Speakers with a cardioid radiation pattern can be placed much closer to the wall behind.
An example is the D&D 8C.
 

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
3,422
Likes
2,407
Location
Sweden
So what about the speaker placed very close to a damped wall, good to damp reflections down to 100 Hz?
 

mixsit

Member
Forum Donor
Joined
Jul 20, 2019
Messages
75
Likes
23
This topic (also called the Hass effect) is covered extensively on the Gearslutz site. Here is one example -
https://www.gearslutz.com/board/stu...nderstanding-precedence-effect-etc-shift.html

You can search for more threads. There are some competent people who frequent that forum. My take away from reading several of these threads is that 1) you need to know what model you are trying to build for your acoustic space (LEDE, RFZ, etc). 2) Recording rooms are different and have different requirements than listening rooms.
I found this post from the link particularly interesting.
https://www.gearslutz.com/board/showpost.php?p=7175004&postcount=120
 

Hipper

Addicted to Fun and Learning
Joined
Jun 16, 2019
Messages
753
Likes
625
Location
Herts., England
If you are serious about hi-fi you should read Floyd Toole's Sound Reproduction third edition.

He discusses the interaction of the room and loudspeakers at great length and it will be discovered that there is a lot going on. For example, you mention the precedence effect and how it impacts the direct sound and it's apparent source. You might think side wall reflections therefore need controlling. But apparently the phantom image created (in your brain, not physically in the room) by stereo has a weakness that may be improved by side wall reflections.

I highly recommend the book (young Floyd posts on here too sometimes).
 

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
3,422
Likes
2,407
Location
Sweden
Reflections can be bad if they are too few and too high in level. I don't think many in this forum dispute that reflections in the listening area that cause a more even diffuse field is bad, especially in stereo listening with its inherent flaws. It also partly masks the timbral change caused by the stereo setup according to Shirley et al (also in Toole's book).

However, I can't see that reflections from the loudspeaker wall is of any benefit. If the wall behind the speaker is acoustically invisible, you will get a better fidelity IMO. I don't thing moving the speaker out from the wall is good either, it will just change the frequency of the interference from speaker-wall reflections. I prefer to have speakers designed for near-wall placement, and use damping panels behind the speakers to reduce the wall reflections.
 

JanRSmit

Member
Joined
Jan 21, 2019
Messages
54
Likes
21
So what about the speaker placed very close to a damped wall, good to damp reflections down to 100 Hz?
Read the reports on dead Chambers( NRC etc). It is even in those rooms difficult to fully absorb the lower frequencies. So there is always some reflections, and if these are more than 2msec later they are not added to the direct sound. Test it yourself.
I have the book of Mr Floyd Toole, it is a very good book. But not an absolute Bible or similar. The 2 msec is an known fact for many, many years.
 

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
3,422
Likes
2,407
Location
Sweden
Read the reports on dead Chambers( NRC etc). It is even in those rooms difficult to fully absorb the lower frequencies. So there is always some reflections, and if these are more than 2msec later they are not added to the direct sound. Test it yourself.
I have the book of Mr Floyd Toole, it is a very good book. But not an absolute Bible or similar. The 2 msec is an known fact for many, many years.

The reflections above 2 ms will be added to the direct sound and cause significant comb filtering that is quite audible. A wall using 10 cm glass fiber and perforated masonite and 6 cm foam damping will be effective down to ca 100 Hz and take care of these effects. The precedence effect will also diminish since the remaining reflections are significantly reduced. Those that are will be added and boost the LF region. I have made those experiments quite many times.
 

Hipper

Addicted to Fun and Learning
Joined
Jun 16, 2019
Messages
753
Likes
625
Location
Herts., England
If you read the link in the OP you will see that:

'The precedence effect appears if the subsequent wave fronts arrive between 2 ms and about 50 ms later than the first wave front. This range is signal dependent. For speech the precedence effect disappears for delays above 50 ms, but for music the precedence effect can also appear for delays of some 100 ms.'

Also:

'If the direct sound is coming from the same direction the listener is facing, the reflection's direction has no significant effect on the results. A reflection with attenuated higher frequencies expands the time span that echo suppression is active. Increased room reverberation time also expands the time span of echo suppression.'

It also seems that the early reflected sound has some directional cues, not enough to be considered echoes but enough to add 'spaciousness':

'A reflection arriving later than 1 ms after the direct sound increases the perceived level and spaciousness (more precisely the perceived width of the sound source).'

I like the phrase 'echo suppression' used in the article but it seems this term usually means something else.
 
Last edited:

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
3,422
Likes
2,407
Location
Sweden
I ageee about the effects when talking about reflections in the live end of the room. You need reflections from side walls etc. The wall behind the speaker is different. Placing the speaker away from the wall to increase reflections > 2 ms would disqualify in-wall speakers and also near-wall speakers with damping behind. The effects of reflections from behind the speaker does not contribute to any positive effects in a traditional speaker setup. The logical is that the listening room is a part of your lounge where the event is happening in front of you. So removing the wall and replacing it with the live scene would be ideal. But since that is not possible, we use speakers and try to create the illusion of that event. Exept for removing the wall, which usually is not possible, you can play with near-wall placement and correct acoustic treatment of that wall.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
I ageee about the effects when talking about reflections in the live end of the room. You need reflections from side walls etc. The wall behind the speaker is different.

It is not different. What is important is timing, not from which wall (side, back, front, ceiling, floor) it is coming. Too much of earl reflections and SQ will suffer.

What is also important to understand is that while room treatment can help too much of it can cause problems as well. 2-4 subs will pretty much alwas help. Room EQ based on spatial measruement is necessary in any case to get linear response.
 

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
3,422
Likes
2,407
Location
Sweden
It is not different. What is important is timing, not from which wall (side, back, front, ceiling, floor) it is coming. Too much of earl reflections and SQ will suffer.

What is also important to understand is that while room treatment can help too much of it can cause problems as well. 2-4 subs will pretty much alwas help. Room EQ based on spatial measruement is necessary in any case to get linear response.

It depends what you mean with different. I’ve done several listening experiments with damping walls and most treatments in the listening area give poor results. I always returned to a ”normal” room with carpet, sofas, pantings on walls etc. The positve effects on SQ have always been correct treatment of the wall behind the speakers. Room EQ is handy for the LF region but not nessesarily needed if you have solved the issues with acoustic methods. As many others I can’t do those things due to other reasons but I’ve listened to such systems and they’ve sounded fabulous.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
It depends what you mean with different. I’ve done several listening exeperiments with damping walls and most treatments in the listening area give poor results.

As long as they have similar timing early reflections from all wals are the same, not different.

Listening experiments are hardly reliable unless they are blind. I suggets you make more measurements to check the difference.
 

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
3,422
Likes
2,407
Location
Sweden
As long as they have similar timing early reflections from all wals are the same, not different.

Listening experiments are hardly reliable unless they are blind. I suggets you make more measurements to check the difference.

I’ve made them and you have seen the effects of RT60 for the simple panels I have now. I’ve had considerably more effective damping walls in the past and I neither can or want to repeat those setups due to other reasons. Its up to others to test and judge for themselves.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
I’ve made them and you have seen the effects of RT60 for the simple panels I have now. I’ve had considerably more effective damping walls in the past and I neither can or want to repeat those setups due to other reasons. Its up to others to test and judge for themselves.

RT60 doesn't make distinction between early and late refletions, if you want to measure that you should apply gating in your measurements.

As you didn't make any measurements of those "more effective dumping" there is unfortunately nothing to judge.
 

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
3,422
Likes
2,407
Location
Sweden
RT60 doesn't make distinction between early and late refletions, if you want to measure that you should apply gating in your measurements.

As you didn't make any measurements of those "more effective dumping" there is unfortunately nothing to judge.

Well I don’t need to make those measurements. The effects are so obvious that anyone treating the backwall would hear it, open or blind. Just take any speaker and measure the effect with and without a backwall to start with. Genelec has published some figures.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
Well I don’t need to make those measurements. The effects are so obvious that anyone treating the backwall would hear it, open or blind. Just take any speaker and measure the effect with and without a backwall to start with. Genelec has published some figures.

Here the folks who usually use the phrase "so obvious no need to measure it" also add that "their wife can hear it while she's still in the kitchen". :D

This is a scientifically oriented forum which means for every claim you make you better provide measurement or a link to some research paper making the same claim.
 
Top Bottom