May i add to Wesayso comment:
for pro world it is usually some acoustic treatment (diffusers) used to introduce some reflections from the front wall in studios. Yes to introduce some reflections to create a sense of spaciousness.
It was a 'hype' at a time (70's/early 80's) but since then most studios who had this kind of treatments (and where principle of acoustic design of control room allowed) just took them out as they tend to 'blur' the image reproduced.
For now it's more about RFZ.
I have a few thoughts....
Much of this paper is about recording techniques and reproduction utilizing a center channel.. Some of it is talking about small room acoustics. But I am having trouble discerning what these points are referring to.
• For speech and most music there are three
time regions:
– early (20-50ms) provides distance
– middle (50-150ms) => distance + confusion
– late (150ms + ) creates reverberance and
envelopment
Ideal Energy Profile
• We need ~-6dB energy in the 15-50ms
range
• flat energy profile to over 160ms
independent in each channel
Over the first read, and earlier in this thread where I posted a graph, I thought he was talking about small room acoustics. But now I am thinking this cant be the case.
1) -6db energy from 15-50ms ??? Really? That would cause reflection havoc wouldn't it?
2) How in the world are you going to preserve >150ms energy in a small room? Add ambient content with a 150ms delay? I haven't seen a DSP yet that allows for this kind of time delay.
So, I am starting to think that these points also apply to recording techniques, not small room acoustics, or more specifically, what one wants a small room acoustic response to be.
Maybe were talking about two different things, but Haas Kickers were NEVER intended to introduce reflections from the front wall. They were meant to work from left and right side/rear (150-110 degrees).
I probably wasn't accurate enough or my memories could be wrong. I'll try to find some more infos about them, maybe i'm wrong and confuse with other things but i'm pretty sure i've seen some being taken out in a control room i used to work in and they were located on l/r walls near the front wall (so the short cut i made in my statement).
I'll check and correct the info if i'm wrong. Thanks pointing that Jim1961.
Ok it was faster than i expected to find the info from 'recording studio design' from Newell: Jim1961 you are right my bad, i must confuse with other treatments which was taken out of the room or both was taken out from the room at same time.
Maybe the one i'm remember was some kind of scattering device.
Maybe the one i'm remember was some kind of scattering device.
I do agree with you though that Haas Kickers have been, for the most part, abandoned in contemporary designs. RFZ is the modern derivation of the earlier LEDE concepts. Problems arise in implementing RFZ in small acoustic spaces though. Most small rooms dont have the space for the QRD's to have the proper distance, and thus, the proper timing to achieve a 20+ms return. Also, the minimum -12db magnitude. So, Kickers are the poor mans termination as opposed to a well designed diffuse return. Its a compromise to be sure. But I have found that having a specular termination is better than none at all. And can sound pretty natural if utilized correctly.
You are right. Some attempt have been done by R.Walker at BBC to manage that for small room. It gave birth to C.I.D. principle. Info about it can be found on BBC R&D white paper (circa 1994/95): a whole study of principle including real world implementation is given. Seems to work quite well given the target (from 1khz up). Still big room for living room 'standard' and not easy to implement for WAF.
A lot in that presentation is about large rooms, but he works his way to how he wanted to implement it into Logic 7 to get a pleasing sound for more than one seat in a small room. JBL came out with the MS-8 for the Car which featured Logic 7. Over on DIYMobileAudio we had Andy Wehmeyer posting who worked for JBL at that time. He had an experimental version of that ms-8 in his own car and even entered some SQ competitions with it. Just to show off the power of good steering algorithms and reverbs in a (very) small room.
So the way I see it, half of that paper is about what we would want and the other half is how to get there in a small space. The reverb would be obvious where it came from as Mr. Griesinger was working for Lexicon at the time
.He even mentions the Lexicon hardware with a "Random Hall" setting.
I have another paper which goes into more detail about the technical part of Logic 7. I'm pretty sure it was linked on this site a while ago.
What I get from that graph is a Haas Kicker at ~20 ms, followed by a period that's more critical (creates confusion) till 150 Hz, and a reverb tail after that. A generated one obviously as we would never get that in a small room. And he makes it clear it should be different in both back left and back right channels.
As you said, it may look like we have a nice tail if one glances at a filtered IR. But the substance of that tail isn't anything resembling a true reverb signal in our small rooms. That's a reason for me to play with software reverb, to try and fill in a little reverb (not too much!) after 150 ms. As that's primarily where envelopment and warmth comes from. At least according to that paper. But it seems I've got to keep the Haas kicker as is, and only try and add a bit of reverb after 150 ms. Keep the rest of the 50 to 150 ms as low as possible.
That's how I read it anyway.
That level of 6 dB was a combined energy of that entire period, 20 to 50 ms. Not sure how to translate it. But for now I'll keep it as a diffused kicker, but no immediate extra's following that kicker in processing, save that for 150 ms and on to create a gradually declining tail with a true reverb nature.
Last edited:
Not much I'd say, but I'm curious enough to try and simulate it to get that "envelopment".
I've heard it to some extend when I used long convolution tails. It "sounds" like a hall at a concert. But that was too much for me. Though part of me liked the effect. Some concert movies will have something similar in 5.1 or higher.
Harman was working on their own version from the looks of that paper.
A lot in that presentation is about large rooms, but he works his way to how he wanted to implement it into Logic 7 to get a pleasing sound for more than one seat in a small room. JBL came out with the MS-8 for the Car which featured Logic 7. Over on DIYMobileAudio we had Andy Wehmeyer posting who worked for JBL at that time. He had an experimental version of that ms-8 in his own car and even entered some SQ competitions with it. Just to show off the power of good steering algorithms and reverbs in a (very) small room.
So the way I see it, half of that paper is about what we would want and the other half is how to get there in a small space. The reverb would be obvious where it came from as Mr. Griesinger was working for Lexicon at the time
He even mentions the Lexicon hardware with a "Random Hall" setting.
I have another paper which goes into more detail about the technical part of Logic 7. I'm pretty sure it was linked on this site a while ago.
What I get from that graph is a Haas Kicker at ~20 ms, followed by a period that's more critical (creates confusion) till 150 Hz, and a reverb tail after that. A generated one obviously as we would never get that in a small room. And he makes it clear it should be different in both back left and back right channels.
As you said, it may look like we have a nice tail if one glances at a filtered IR. But the substance of that tail isn't anything resembling a true reverb signal in our small rooms. That's a reason for me to play with software reverb, to try and fill in a little reverb (not too much!) after 150 ms. As that's primarily where envelopment and warmth comes from. At least according to that paper. But it seems I've got to keep the Haas kicker as is, and only try and add a bit of reverb after 150 ms. Keep the rest of the 50 to 150 ms as low as possible.
That's how I read it anyway.
That level of 6 dB was a combined energy of that entire period, 20 to 50 ms. Not sure how to translate it. But for now I'll keep it as a diffused kicker, but no immediate extra's following that kicker in processing, save that for 150 ms and on to create a gradually declining tail with a true reverb nature.
You know a lot more about the DSP aspect of what he is talking about than I. So I value your opinion, especially in this area.
While I see your points, I am still not sure. In a sense, he seems to be trying to derive what different halls sound like, measure like in a reverb sense, and apply that to a listening room. If so, I have to ask myself if I want my room to sound like those halls. That might work well for classical, but hard to imagine it working for progressive metal. The other point is that the recordings we listen to already have the delays and reverb as part of it. Trying to make a room add those features again seems like double dipping the ambiance, in a sense.
I also find myself going back to something you said. That is, you dont want to hear the room. Following the prescription Mr. Griesinger is laying out, certainly one would be bringing the room more into the equation. And it seems to me, the >150ms energy, as he stated himself, is treated by the mind as a separate event, and thus would have to perceptually construed as a added event.
Sorry if I sound a little critical. Its just how I think. I need to be able to see in my mind all the involved aspects of an idea before I can wrap my head around it. And in Mr. Griesinger's paper, I see a lot left out and open to guesswork.
For instance:
1) While he shows ETC reverb graphs, there is no legend to say what levels of magnitude to shoot for. He describes the 15-50ms as -6db, but makes no mention of what the medium (50-150ms) and late (150ms+) energy levels should be.
2) He also doesn't describe the spectra. What should the FR curves look like at these various delayed levels?
3) I also get no sense of the directional elements. How much from the front, back, and sides and so on.
Hopefully its obvious I am interested in this research, or I wouldn't be spending time trying to decimate it.
As am I, I've read much of this stuff before, to me it made sense to hide my room by superimposing another, better room, but I would only want to do that to forget about that room, not add a distinct echo. But I do know the feel of floating energy around you in a good concert hall. In my mind I am looking for just a hint of that... At the lowest level possible to still try and hide what's really there (the small room, mine actually).
If you attend a concert, you must have noticed one room/hall sounding way more pleasant than others. I want the pleasantness, but not at the cost of it being obvious.
You're absolutely right in the fact that most recordings already have an ambience. But in two channel listening I sense it defines the front part of the room. Not creating that bigger listening space. The Haas kicker helps, to me it creates a more lively 3D feel. But that sense of energy floating around you, you can sense at a concert isn't there.
When I didn't have my damping panel behind me, I had strong reflections off of that wall. That gave a sense of envelopment, but only at higher frequencies. It also muddied the imaging. But that feeling of envelopment is a pleasant one. At least to me. If I can get a hint of that, short and low enough in level not to be too obvious... worth a shot? It will be way down in level, but it's created from the actual music. Similar to how you can hear drum hits bounce off the walls in recordings. This time creating the back part of the room. At least that's how I imagine it in my mind. My only worry is: will it be too much?
After listening to reverb tails on headphones I wasn't completely convinced yet. But taking out the middle part seemed to help. More experimenting to be done...
Too bad your MiniDSP can't run VST plugins, can it? Although they probably have their own reverb plugins?
There are so many choices and as many questions on this right now...
A few answers were there in this paper, I'll hunt down that other one.
I've looked into this before, a while ago... it showed that compared to other multi channel devises Logic 7 often gets praise for it's 2 channel to surround capabilities. I'm just not sure I want to go that far. Lot's of the same doubts you seem to have. But without trying, how would I know?
Even Linkwitz is using an AVR with some type of mix down in the page I linked from him. But I do know he's not listening to the same stuff I am.
The kicker is here to stay for me. This reverb experiment is just that, an experiment.
If it does something useful? It could become a more permanent part of my processing.
But right now only my curiosity wins. But I have doubts...
Edit: uploaded the technical document I saved a while ago. Pretty sure I got it here but I wouldn't want to violate anything by attaching it.
If you attend a concert, you must have noticed one room/hall sounding way more pleasant than others. I want the pleasantness, but not at the cost of it being obvious.
You're absolutely right in the fact that most recordings already have an ambience. But in two channel listening I sense it defines the front part of the room. Not creating that bigger listening space. The Haas kicker helps, to me it creates a more lively 3D feel. But that sense of energy floating around you, you can sense at a concert isn't there.
When I didn't have my damping panel behind me, I had strong reflections off of that wall. That gave a sense of envelopment, but only at higher frequencies. It also muddied the imaging. But that feeling of envelopment is a pleasant one. At least to me. If I can get a hint of that, short and low enough in level not to be too obvious... worth a shot? It will be way down in level, but it's created from the actual music. Similar to how you can hear drum hits bounce off the walls in recordings. This time creating the back part of the room. At least that's how I imagine it in my mind. My only worry is: will it be too much?
After listening to reverb tails on headphones I wasn't completely convinced yet. But taking out the middle part seemed to help. More experimenting to be done...
Too bad your MiniDSP can't run VST plugins, can it? Although they probably have their own reverb plugins?
There are so many choices and as many questions on this right now...
A few answers were there in this paper, I'll hunt down that other one.
I've looked into this before, a while ago... it showed that compared to other multi channel devises Logic 7 often gets praise for it's 2 channel to surround capabilities. I'm just not sure I want to go that far. Lot's of the same doubts you seem to have. But without trying, how would I know?
Even Linkwitz is using an AVR with some type of mix down in the page I linked from him. But I do know he's not listening to the same stuff I am.
The kicker is here to stay for me. This reverb experiment is just that, an experiment.
If it does something useful? It could become a more permanent part of my processing.
But right now only my curiosity wins. But I have doubts...
Edit: uploaded the technical document I saved a while ago. Pretty sure I got it here but I wouldn't want to violate anything by attaching it.
Last edited:
A lot of my decision process on what to try and what to not boils down to simplicity. A Haas kicker can be as simple as leaning two boards against a wall. Way, way too simple not to try. But if I have to spend tons of money, it better make sense and attract a lot of praise from people I respect.
As far as reverb restoration in small listening rooms goes, it seems everyone has a different idea of going about it and no clear consensus from my piers on any one of them. Even the available measurement data differs considerably.
So while I am interested, I am only in research mode. I am not jumping into anything complex right now unless its really simple.
As it turns out, my ambient project cost me a total of $100 US dollars. My miniDSP was $100. These are the kind of upgrades that I will jump into easily.
As far as reverb restoration in small listening rooms goes, it seems everyone has a different idea of going about it and no clear consensus from my piers on any one of them. Even the available measurement data differs considerably.
So while I am interested, I am only in research mode. I am not jumping into anything complex right now unless its really simple.
As it turns out, my ambient project cost me a total of $100 US dollars. My miniDSP was $100. These are the kind of upgrades that I will jump into easily.
Perfectly understandable...
For me it is as much to learn something as it is curiosity. I seem to have the opportunity as I have the DSP power at hand. Still it's a puzzle to get it right I assume. I was intrigued by what I've read a long time ago on the Car audio forums. Yet, the JBL MS-8 didn't quite make it into permanent setups of avid Car audio sound quality enthusiasts. Most just played with it and moved on, at least that was my impression.
This whole concept is a bit double for me. I've spend a lot of thought, time and effort in getting/maximizing the ambience present in the recording. So this really feels like adding something that doesn't belong in that concept. But as long as I haven't tried it I cannot really judge that. And who's to say my home brew kicker is according to the (unwritten) rules. It's just a few steps up from the old ambient tricks of the 70's. But it does seem to do something useful for the perception.
It won't cost me anything, except when I decide it's worth it to stay. A demo plugin is only going to bring me the opportunity to try. The levels needed will be hard to figure out. And the timing too. If it doesn't work out I'll only lose time. But maybe, just maybe I'll learn something again.
For me it is as much to learn something as it is curiosity. I seem to have the opportunity as I have the DSP power at hand. Still it's a puzzle to get it right I assume. I was intrigued by what I've read a long time ago on the Car audio forums. Yet, the JBL MS-8 didn't quite make it into permanent setups of avid Car audio sound quality enthusiasts. Most just played with it and moved on, at least that was my impression.
This whole concept is a bit double for me. I've spend a lot of thought, time and effort in getting/maximizing the ambience present in the recording. So this really feels like adding something that doesn't belong in that concept. But as long as I haven't tried it I cannot really judge that. And who's to say my home brew kicker is according to the (unwritten) rules. It's just a few steps up from the old ambient tricks of the 70's. But it does seem to do something useful for the perception.
It won't cost me anything, except when I decide it's worth it to stay. A demo plugin is only going to bring me the opportunity to try. The levels needed will be hard to figure out. And the timing too. If it doesn't work out I'll only lose time. But maybe, just maybe I'll learn something again.
I left that out of my rant. But I completely agree that this too is a major reason for my continued interest.
No conclusions yet, I did have a lot of fun testing some reverb settings. Much more work is needed though. I checked the timing on my "Haas kicker". I suffer from the same problem you are Jim, my HF is late to the party (bouncing off the environment, while the low end is more direct).
A couple of all-pass filters helped but isn't my permanent solution.
Here's a quick graph:
I guess I can get a bit of reverb to last... The comparison graph is without Haas kicker by the way. I lowered my usual Haas Kicker and extended the reverb with a Hall reverb, shortened to ~1 second from it's original ~3.5 second length. The early part is reduced in level and the late reverb (hopefully after 150 ms, but that seems valid judging the graph) turned up a bit.
It's a longer tail, mainly kicking in after 150 ms. I had a lot of fun listening to this. But I need way more time to get it right. This was just a quick test. Those are never a good idea.
A couple of all-pass filters helped but isn't my permanent solution.
Here's a quick graph:
I guess I can get a bit of reverb to last... The comparison graph is without Haas kicker by the way. I lowered my usual Haas Kicker and extended the reverb with a Hall reverb, shortened to ~1 second from it's original ~3.5 second length. The early part is reduced in level and the late reverb (hopefully after 150 ms, but that seems valid judging the graph) turned up a bit.
It's a longer tail, mainly kicking in after 150 ms. I had a lot of fun listening to this. But I need way more time to get it right. This was just a quick test. Those are never a good idea
.
The beauty of processing is I can use a reverb of a large room and shorten it's reverb tail. But it will still have the properties of that large room. It gently rolls off the top end.
But as each parameter can be changed, it will take me a while to find optimum settings.
Judging the reverb alone, from the RT tab that envelopment region is exactly what I am targeting. Useful graph! Thanks for posting that. I'll have to work on my Haas kicker first, then try to optimize the reverb. It did show promise! I loved what I heard. But I've got to love it over a long time to be convinced.
But as each parameter can be changed, it will take me a while to find optimum settings.
Judging the reverb alone, from the RT tab that envelopment region is exactly what I am targeting. Useful graph! Thanks for posting that. I'll have to work on my Haas kicker first, then try to optimize the reverb. It did show promise! I loved what I heard. But I've got to love it over a long time to be convinced.
And the one above 1khz are for sharp image/stereo.
More than the high roll off it is the way the hall algorythm deal with the 'space' between reflections creating reverbs ( a bit like the spread of mode in the low end region: more spaced=more enjoyable) and the way they are 'spread' over the room to be simulated.
Jim1961: you've done an impressive work with your own room.
Last edited: