LEDE / RFZ ideas applied to a critical listening room

[jim1961]

Originally Posted by bwo

I was thinking sitting closer to the frontwall then the middle. The middle is often a place with most nulls, so we don't want that. I don't see why the speakers need to be so far out from the wall. Basstraps can take care of resonances. But you know what you can do. Right now the ISD gap is too short.

If i am understanding the pic (model) correctly, the rear diffuse energy bounces off the lateral walls to create the ISD termination. So it would seem you would either need angled walls or large deflection panels to achieve this result in this way.

Basically you don't want diffusers to diffuse energy towards other diffusers. When the energy is diffused, there's no need to diffuse it again. I think there would be better ways to place them. But again, depends on what you can do.
Placement of diffusers depends on the topology of the room and the nature of the speakers. Areas like levels, arrival times and spatial temporal distribution of the later arriving diffuse field are what we look for.

This is one layout, but not necessarily something you could copy. It depends on several factors like mentioned above. RPG are diffusors.

The pic still looks like a control room model to me. Remember, this is a living room. I have people over. I dont want the listening area to be squonched up into the front third of the room. So for this, and other reasons, I really need to keep the listening position where it is. And that basically dictates where the speakers must be.

If you believe 10ms is too short a ISD gap, then I maybe out of luck. This is a point regarding the LEDE / RFZ that many disagree about and/or interpret differently. In some instances, I see quotes referring to only the first 6ms as being critical to suppress leading some to believe that means a ISD termination can be 6ms also. I have posed the 10ms question on other forums and some people think its fine, while a few feel as you do. Personally, I am not sure what to believe. But I have tried to move the listening chair forward (keeping the rest the same except re-angling the lateral diffusers) and basically what happens is the direct signal path shrinks making the ISD termination more like -14 or -15db.

But your point about the flanking and rear diffuser locations may be a good one. The lateral diffusers have a obvious function and purpose. But the others are just kinda there to add to liveliness.

btw, I do appreciate your input. Thanks.

edit: I did want to add that I do understand why you are suggesting to move things forward. I do also see how I could lengthen the ISD gap by doing so. But it seems to me that to make that work, I would have to emulate the example pic you posted. Angling the walls for instance isnt something I am willing to do. And I have those three hallway / openings to other rooms that I cant change either. So there are some constraints here. Basically, I just want to make the most of what I can within my constraints.

[Jens Eklund]

Originally Posted by jim1961

1) Keep reflections to -20db or better up to the ISD termination.
2) Make the ISD termination energy around -12db from direct signal.
3) Make the ISD gap between 10-25ms
4) Let the termination energy arrive from a lateral perspective.

Do I have all this right?

1. Yes.

2. About -12 dB or higher is preferable in order to reach the threshold of Haas.

3. The ISD-gap should be determined by the recording room and the control room should have an ISD-gap of at least 3-5 ms longer than the recording space. If there is no dedicated recording space, the ISD-gap should be as long as possible (up to about 25-30 ms) in order to accurately hear the acoustic print of various spaces, big or small. 10 ms is on the limit of too short in my opinion. I would consider excluding the termination if the ISD-gap is that short. About 12-15 ms is i m o a practical limit of ISD-gap lengths and 20-25 ms being the ideal.

4. Yes, as much as possibly but you naturally will also see some energy from the rear. Also important is that the slope of the decaying (semi) diffuse field should be exponentially decaying (linear on a log dB scale) without large irregularities or sparse higher than average reflections.

[jim1961]

Originally Posted by Jens Eklund

3. The ISD-gap should be determined by the recording room and the control room should have an ISD-gap of at least 3-5 ms longer than the recording space. If there is no dedicated recording space, the ISD-gap should be as long as possible (up to about 25-30 ms) in order to accurately hear the acoustic print of various spaces, big or small. 10 ms is on the limit of too short in my opinion. I would consider excluding the termination if the ISD-gap is that short. About 12-15 ms is i m o a practical limit of ISD-gap lengths and 20-25 ms being the ideal.

One idea I am tinkering with is putting deflection panels at the rear redistributing that energy to the lateral QRD's I have. Working out the time and distances would let that energy get to the listening position at around 22ms. But I am not confident I can achieve the -12db termination threshold in this configuration. If I deflect ALL of the rear energy to the sides, I might get a -15 or -14db termination.

So, which is better.

1) 10ms gap, -12db termination
2) 22ms gap -15db termination
3) No termination.

????

And if no termination, I am still looking to keep the RFZ right?

[Jens Eklund]

Originally Posted by jim1961

One idea I am tinkering with is putting deflection panels at the rear redistributing that energy to the lateral QRD's I have. Working out the time and distances would let that energy get to the listening position at around 22ms. But I am not confident I can achieve the -12db termination threshold in this configuration. If I deflect ALL of the rear energy to the sides, I might get a -15 or -14db termination.

So, which is better.

1) 10ms gap, -12db termination
2) 22ms gap -15db termination
3) No termination.

????

And if no termination, I am still looking to keep the RFZ right?

Not exactly sure what you mean but I would favor the second option assuming the entire ISD-gap is at least -20 dB down.

[jim1961]

Originally Posted by Jens Eklund

Not exactly sure what you mean but I would favor the second option assuming the entire ISD-gap is at least -20 dB down.

So a weak termination still has some effect? I was under the impression that if you didnt reach the threshold of -10 to -12db, you didnt have a ISD termination at all.

Or do you agree with me on this point (that -15db isnt really a ISD termination) but you feel it would overall just sound better than a well terminated short (10ms) ISD?

[gigi]

Originally Posted by jim1961

Remember, this is a living room. I have people over. I dont want the listening area to be squonched up into the front third of the room. So for this, and other reasons, I really need to keep the listening position where it is. And that basically dictates where the speakers must be.

... then you should probably remove the "critical" out of the name of this thread and out of your expectation.

my 2c

[jim1961]

Originally Posted by gigi

... then you should probably remove the "critical" out of the name of this thread and out of your expectation.

my 2c

I can critically listen to a AM radio. Aiming for perfection of an idea isnt the same as expecting to attain it. My expectations are not more, or less, than what CAN be achieved within my constraints.

I have exceeded my expectations on numerous occasions already when it comes to the quality of the sound in my room. Thus, I dont yet know what the upper limit is or how to achieve it.

But as to your response. I intend to listen to music critically in this room. This is independent of what it ultimately sounds like.

[Mctwins]

Hi
Show some measurements of your room?

[Mctwins]

and also...
Why is it so important for you, that you have decided to use LEDE/RFZ concept in your livingroom?

[OpusOfTrolls]

Are those wood palettes?

[bwo]

Originally Posted by jim1961

edit: I did want to add that I do understand why you are suggesting to move things forward. I do also see how I could lengthen the ISD gap by doing so. But it seems to me that to make that work, I would have to emulate the example pic you posted. Angling the walls for instance isnt something I am willing to do. And I have those three hallway / openings to other rooms that I cant change either. So there are some constraints here. Basically, I just want to make the most of what I can within my constraints.

I never said you should copy it......
It was simply one layout. Like mentioned before, the placement of diffusors will depend on the topology of the room and how the speakers disperse. Ideally the diffusion should arrive laterally, but that doesn't mean that the diffusors needs to be placed on rear sidewalls. You can angle the diffusors on the rear like a V or an inverted V, so they will send most of the energy towards rear sides. Today, you have some double diffusion going on though. The ones in the back are sending some energy towards the ones on the side of the listening position. I would remove the ones on the sides of the listening position and put more farther back. What's the idea with the deflectors? I would assume they are partially in the way of the speakers.

If you can't more the setup forward, I wouldn't use a termination of the ISD gap. Maybe you could use 2D diffusors that will lower the energy more then 1Ds. That might be better then only absorption.

A strong termination of the ISD gap (minimum of -12dB) isn't very vital with LEDE/RFZ. That's something you can get away with, but you one need to pay more attention then to the slope of exponentially decaying diffuse field. In your situation, a longer ISD gap is much more important. If that's not possible, you would be better attenuation the arriving energy at 10 ms IMO.

[jim1961]

Originally Posted by Mctwins

Hi
Show some measurements of your room?

[jim1961]

Originally Posted by Mctwins

and also...
Why is it so important for you, that you have decided to use LEDE/RFZ concept in your livingroom?

Its just a place to start. Something to try. I want to know what this model sounds like when optimized. If in the end, I dont like it, I will try something else. But right now, I am still in the optimization phase.

[jim1961]

Originally Posted by bwo

What's the idea with the deflectors? I would assume they are partially in the way of the speakers.

The Partition / Deflector attached to the side walls keeps energy from the right speaker out of the right diffusion panels (and visa versa with the left channel). Those energies were arriving at 7-8ms, so the partition blocks those to protect the RFZ.

Originally Posted by bwo

A strong termination of the ISD gap (minimum of -12dB) isn't very vital with LEDE/RFZ. That's something you can get away with, but you one need to pay more attention then to the slope of exponentially decaying diffuse field. In your situation, a longer ISD gap is much more important. If that's not possible, you would be better attenuation the arriving energy at 10 ms IMO.

Lowering the energy at 10ms is easy. I thought a strong termination was VITAL to LEDE /RFZ ?

Perhaps what you are saying is a weak termination or non at all is better in my case because of how early it arrives?

This seems to be what Jens is saying also.

[Mctwins]

Originally Posted by jim1961

Its just a place to start. Something to try. I want to know what this model sounds like when optimized. If in the end, I dont like it, I will try something else. But right now, I am still in the optimization phase.

Ok, I understand.

About the measurement, I would like to see the freq response and the waterfall plot.

[bwo]

Originally Posted by jim1961

Lowering the energy at 10ms is easy. I thought a strong termination was VITAL to LEDE /RFZ ?

Not crucial. You can get away with it, but you will need to pay more attention to later high gain reflections. The slope of diffuse filed becomes more important to pay attention to when you don't have a Haas kicker.

Originally Posted by jim1961

Perhaps what you are saying is a weak termination or non at all is better in my case because of how early it arrives?

I believe 10 ms is too short for a LEDE/RFZ concept. So yes, I wouldn't strive for strong termination. I would either attenuate (probably with 2d diffusors) or move the setup.

[localhost127]

Originally Posted by Mctwins

Ok, I understand.

About the measurement, I would like to see the freq response and the waterfall plot.

why. neither is relevant to the context of the discussion as generated by the OP.

[jim1961]

Assuming I can track down the peaks at 35ms, is this more in order with what some of you are trying to steer me towards?

[jim1961]

Nevermind, that 22ms plan just wont work.



I went back to the old setup, changed a couple things and managed 1 more ms. Now termination is at 11ms. Without a complete overhaul of my room, or complete new understanding of how to do it, I cant see how to improve upon this for now.

[aackthpt]

I'd think you'd want to achieve an even longer ISD gap than for the typical control room given that you aren't monitoring for an environment with a known ISD gap. As such I'd think a longer gap would be better than a higher termination, in the interests of enforcing the characteristics of the recording more than your room. The limit of that thought process is effectively anechoic, which is a losing battle as a comfortable living room (though I'd be super intrigued to walk into a non-environment living room LOL). Having said that, and taking into account the article summary in and linked from this post (which seems to indicate that different room responses are desirable for different kinds of music), I think the only reasonable way out would be to try to go ambechoic. Or at least some sort reasonable and possible facsimile thereof. So perhaps get rid of the deflectors, attack all early arrivals with 2D diffusors (including ceiling bounce), if the levels are still too high play with deflectors or strategic absorption ("perfect is the enemy of the good"--mark rippetoe), add more diffusers to get the decay even. Assuming you like the sound, the next step is to try to avoid the upgradeitis for as many years as possible, and simply listen to music.

And if you really want a realistic music experience, go check it out in real. Most of us in 'murca could probably stand to get out of our living rooms a bit more.

Personally, I used to think of control room acoustics as a great idea for a listening room. More study and I think they're different animals and probably shouldn't be the same in general, particularly for anyone who has wide-ranging musical tastes. Some of the concepts have crossover, but taking a control room model literally doesn't seem advisable--in part because it ignores individual preferences and self-imposed limitations (such as in your case).

So, OP, have you read the white paper on listening rooms in this thread? I'm sure those gents understand the LEDE/RFZ response and have good reasons for concentrating on other factors first. In the paper it's not as though they hold back technically to try to appeal to lay people. Also, Jeff addressed the response model somewhat in his final response to the thread. The viewpoints he expressed in that post may contain significant wisdom in this area -- and I think may leave the door open to a variety of response models working well.

Regards,
John

[jim1961]

Originally Posted by aackthpt

I'd think you'd want to achieve an even longer ISD gap than for the typical control room given that you aren't monitoring for an environment with a known ISD gap. As such I'd think a longer gap would be better than a higher termination, in the interests of enforcing the characteristics of the recording more than your room. The limit of that thought process is effectively anechoic, which is a losing battle as a comfortable living room (though I'd be super intrigued to walk into a non-environment living room LOL). Having said that, and taking into account the article summary in and linked from this post (which seems to indicate that different room responses are desirable for different kinds of music), I think the only reasonable way out would be to try to go ambechoic. Or at least some sort reasonable and possible facsimile thereof. So perhaps get rid of the deflectors, attack all early arrivals with 2D diffusors (including ceiling bounce), if the levels are still too high play with deflectors or strategic absorption ("perfect is the enemy of the good"--mark rippetoe), add more diffusers to get the decay even. Assuming you like the sound, the next step is to try to avoid the upgradeitis for as many years as possible, and simply listen to music.

And if you really want a realistic music experience, go check it out in real. Most of us in 'murca could probably stand to get out of our living rooms a bit more.

Personally, I used to think of control room acoustics as a great idea for a listening room. More study and I think they're different animals and probably shouldn't be the same in general, particularly for anyone who has wide-ranging musical tastes. Some of the concepts have crossover, but taking a control room model literally doesn't seem advisable--in part because it ignores individual preferences and self-imposed limitations (such as in your case).

So, OP, have you read the white paper on listening rooms in this thread? I'm sure those gents understand the LEDE/RFZ response and have good reasons for concentrating on other factors first. In the paper it's not as though they hold back technically to try to appeal to lay people. Also, Jeff addressed the response model somewhat in his final response to the thread. The viewpoints he expressed in that post may contain significant wisdom in this area -- and I think may leave the door open to a variety of response models working well.

Regards,
John

I suppose where I am at is determining which of these three ideas is the most important:

1) ISD gap (10-25ms)
2) Direction of termination (all say lateral)
3) Strength of termination (-12db?)

I can probably have 2 of those 3, but not all 3.

So, for all those that continue to follow this thread, which 2 would you choose if you couldnt have all 3?

---------------- --------------- -----------------

John, I have read that white paper and actually have conversed with Nyal some. Certainly, some of the points made in that paper I am taking into account, namely tracking the ETC at different frequencies.



Red = 668hz
Orange = 1k
Yellow = 1.6k
Green = 2,6k
Purple = 4.2k

(all 1 octave bandwidth)

[jim1961]

I know you guys mean more than this, but ive got it to 11.6ms. Still laterally arriving and still -12db termination.

[aackthpt]

Your earlier curves had the returns at 2.5ms lower, in some cases much lower (post 13). What happened?

How does the sound of this (post 23) setup compare to the post 19 setup?

[jim1961]

Originally Posted by aackthpt

Your earlier curves had the returns at 2.5ms lower, in some cases much lower (post 13). What happened?

How does the sound of this (post 23) setup compare to the post 19 setup?

That 2.5ms hump is a floor reflection. It is troublesome and varies with different alignments. Ignore it for now because I havent given my full attention to it yet.

The post 19 setup I never even listened to given the 35ms blast.

The post 23 setup sounded a trifle better than where I started (10ms ISD).

But ive been tinkering some more and have a new setup. (see next post)

[jim1961]

14ms anyone ?

[bwo]

Why not the one with 25 ms and trace down the 35 ms bumb?

[jim1961]

Originally Posted by bwo

Why not the one with 25 ms and trace down the 35 ms bumb?

Looking at just the data, I can see where that one would seem the bread winner, but it had unsolvable other consequences.

The same spot responsible for the 25ms termination was also responsible for the 35ms blast. A double reflection. It was either keep both or get rid of both. Sad really cause on paper (computer graphics), that one looked real nice minus the 35ms blast. I played with it for 3 hours and just couldnt resolve it.

also, because the termination in that configuration was based on bouncing of off deflectors and then into QRD's, it took all the energy of the room to get the termination. There was nothing left for any kind of diffuse field to follow.

My newest 14ms ISD termination model is derived from direct first reflections off of QRD's, so I have more room energy to work with and have a better chance of a diffuse field to follow. I suppose you may feel 14ms is still too short?

[jim1961]

Originally Posted by jim1961

Attachment 306875

14ms anyone ?

This is sounding real nice. I can actually hear more of the recording ambiance than before. The ambiance (reflections) sound more natural, less artificially room contrived.

[bwo]

You on the right track.
Remember to also compare the frequency response and waterfall. A longer ISD gap alone isn't necessarily going to sound better if the other become problematic.