Looking up references to answer a question posed over at Ethan's EQ forum, I happened to take a fresh look at Gilford's 1979 AES paper, The Acoustic Design of Talks Studios and Listening Room. (This paper was originally presented for the Radio and Telecommunication Section of the IEEE.) I wanted to quote a few sections that have relevance here:
Gilford wrote:
There will be room modes...in all parts of the audible spectrum, but whether or not they will be appreciable as colorations depends on the following factors:
(1) the bandwidth of the mode
(2) the degree of excitation of the mode
(3) its separation from neighboring strongly excited modes
(4) the positions of the sound source and microphone with respect to standing-wave systems
(5) the frequency content of the source.
What's interesting is that the above was detailed in Section 2.1, Formation of Simple Modes in Small Rooms, that discussed only axial modes.
He goes on (from Section 2.3 The Relative Importance of Axial, Tangential, and Oblique Modes):
Gilford wrote:
For the case of a typical small studio...calculation shows that no frequency is likely to become prominent unless it (has) a high early intensity and a long decay. This condition* is satisfied only by the axial modes, which are therefore the only ones likely to become individually significant. An exception to this rule is that a few tangential or oblique modes of low frequency may possibly be audible, owing to their high initial intensities or wide spacings.
Emphasis on the last sentence was added by me.
{*Gilford is explaining an image-source method of calculation. I omitted references to it to keep things simple here.}
Finally, I believe I may have discovered another alternative for Ethan to consider for his axial mode calculator:
Gilford (context: from Section 3.4 Design Precautions for Avoidance of Colorations) wrote:
A simple calculation...enables a list of all the axial modes for all three dimensions to be written down in order of frequency. It will be unnecessary to continue the list beyond, say, 350 Hz because...the axial modes in a well-designed talks studio will not be prominent above that frequency. The list must next be examined to find modes, or groups of modes with almost the same frequency, which are separated from their nearest neighbors on either side by intervals appreciably larger than their bandwidths. In practice the minimum separation for audibility appears to be about 20 Hz.
Modes or groups separated from their neighbors by greater intervals than this should be noted, and...attempts should be made to alter the groupings by changes in the proposed room dimensions.
IMO, the above gives Ethan the option of adding some sort of mode spacing evaluator. Spacings wider than 20 Hz can be called to the attention of the user and steps can be taken then to change the "proposed room dimensions" to give a better spread of axial modes.
The science makes sense. And it takes the "need" for calculating "non-axials" out of the picture.
Whaddyathink, Ethan?
{Note: Posting same message in other thread.}
_________________ ---lovecow---
"You've got to be very careful if you don't know where you're going, because you might not get there." - Yogi Berra
> This must be the thread that you are referring to at RSD. <
The very fact that you bring it up here shows clearly your intent to continue making trouble. If you have nothing useful to contribute to the science of acoustics, why are you even posting?
--Ethan
Ethan:
You are wrong!
Andre linked that thread here, because, contrary to your claims, it quite clearly shows Steve getting annoyed at you for posting in your typical defensive and agressive style. He may have asked eveyrone to cool it, but it was you who started the fire there.
You must be on auto-pilot, as it only takes an ounce of rational to realise it's pointless posting claims that can so easily be disproved.
Andre was just putting the record straight. I don't blame him when he's been wrongly accused.
You are always accusing people of causing trouble, so I can only assume your blind to the fact that those people are only trying to put the record straight, after you started the trouble in the first place.
It's a bit like a burglar accusing a homeowner of starting trouble when he defends himself.
I offered an Olive branch to try and sensibly talk about this, and hopefully calm things down, and eventually end this stupid feud.
Paul
NOTE: this post is NOT an attack, but an observation.
In the "Master Handbook Of Acoustics 4th" pg 343, it says "Gilford states his opinion that an axial mode separated by more than 20hz from the next axial mode will tend to be isolated acoustically. It will tend not to be excited through coupling due to overlapping skirts but will tend to act independantly. In this isolated state it can respohnd to a component of the signal near its own frequency and give this component its proportional resonant boost."
The reference/bibliography says that is from what looks like the same place you found, except ultimately it was from an earlier print, namely: Gilford, C.L.S. "The Acoustic Design of Talks Studios and Listening Rooms", Proc. Inst. Elect. Engs., 106, Part B, 27 (May 1959) 245-258. Reprinted in J. Audio Eng. Soc., 27, 1/2 (1979) 17-31
But what concerns me is the last part of Gilford's
Quote:
In practice the minimum separation for audibility appears to be about 20 Hz.
If I play three notes together (25.5hz, 32.7hz, 41.2hz) what I hear is a chord (A major I think). Each of those three has a separation well less than 20hz.
And looking at a modal calc for a few rooms (e.g. 12.46, 11.42, 7.90ft), there are no two frequencies separated by 20hz. I don't think that Girlford meant that there are no rooms with modal problems.
So I'm thinking that Gilford's 20hz might be taken in context of his mentioning 350hz.
Again, on a piano, F @ 349hz, the next semitone down is E @ 330hz, which is roughly 20hz difference. In other words, the "separation for audibility" is a semitone.
I think it's an axium that people can hear the difference in tone of any two different semitones.
A semitone, by interesting coincidence, is roughly about as Bonello's rule of 5% of their frequency apart. i.e. 349hz/330hz = 1.0575757 or a difference of about 5%.
Perhaps people can hear differnces less than a semitone as well.
Here is an interesting summary of how the ear works
but more to the point is this experiment in how close two frequencies can be heard. Based on the forumula there, and assuming 7 cents (cent = 1/1200 octave) is the minimum audible frequency sparation, for a 1hz difference that works out to 250hz vs 251hz is a 7 cent difference. I won't bother with the 3.5 cent and 5 cent differences mentioned in the conclusion for no particularly good reason of mine, simply because obviously differences this small are no where near the order of magnitude larger scale that Gilford and Bonello are talking about, so I'll try with the 5% rule for my calcs, i.e. frequencies within 5% and separated by 5% from others, may be bad modally.
5% = 85 cents = 0.0708 of an octave.
Quote:
IMO, the above gives Ethan the option of adding some sort of mode spacing evaluator
It's been my plan to do that with my little webpage. I just have to find the time...
proudtower Recording Org Pro Audio Forums
Joined: Mar 14, 2005
Posts: 145
Location: achterhoek, netherlands
If I understand what you mean, this reffers to Ethans reply on my comment that human sound perception in the low frequencies is more lineair, and in higher frequencies (above 200 Hz, I think) is logaritmic.
(I probably should come up with a reference, but I have to search).
Ethan says that rearranging the frets on his bass does make a difference. I agree with him, but that is not the point.
The 20 Hz seperation doesn't mean you cannot hear other seperations or scales, but that spacings more than 20 Hz apart are annoying. And modes spaced some distance on the freq scale in practice can/will have larger level differences than in theory, where they are equal or differ only 3 or 6 dB.
Personally I don't care to much about modes (Jeezzz, what am I saying now!).
When you have a not to big room you cannot sacrifice space to make the room dimension ratio perfect. You have to do with what you have. Meaning that experimenting with damping and speaker/listening position has to do the trick, and it almost always will.
Bert
eric_desart Recording Org Pro Audio Forums
Joined: May 23, 2003
Posts: 816
Location: Belgium Antwerp
But I also believe Bert, that you're more busy with listening spaces than recording spaces.
Our brain solves a lot.
Recording spaces implies more accuracy.
But of course experience can solve a lot to. But you also have a trained instinct well aware of the concept modes and interference.
_________________ Best Regards - Eric Desart
Also or Non Respected Past RO Acoustics & Design Moderator?
proudtower Recording Org Pro Audio Forums
Joined: Mar 14, 2005
Posts: 145
Location: achterhoek, netherlands
I'm with Bert. I.e., I think the "20 Hz" thing is not about what is audible in terms of pitch. Rather, it's about what is perceptible in terms of being (subjectively) problematic. I interpret Gilford's statement to mean that spacings greater than 20 Hz give a distinct, unwanted character to the room - in his terms, the "coloration" of the sound. Spacings less than about 20 Hz, he seems to believe, are less inclined to cause these big colorations. From a modal bandwidth standpoint, this does make sense to a degree. Modal bandwidths are typically 3 to 10 Hz (according to Bonello - Everest gives a slightly wider range, I think). Therefore, if you have wide axial mode spacings, the 20 Hz criteria lends itself some creedence.
Hypothetically: Two modes, 60 and 75 Hz. Assume a bandwidth of 10 Hz. Generally, this means frequencies between 55 and 65 Hz will excite the former and frequencies between 70 and 80 Hz will excite the latter. What Gilford is saying is that these ranges are so close, their response curves tend to overlap enough to smooth out the colorations. Now consider two modes at 60 and 85 Hz, Gilford is saying that the gap between 65 Hz (the top of the first mode's bandwidth) and 80 Hz (the bottom of the second modes bandwidth) is not sufficiently small to smooth the colorations.
At least that's how I interpret it. If you listen to LF sound in a "cubic" room, you get a wide variation of build-up and (what many of my customers like to refer to as) "suck-out." A room with a better spread of room dimensions (ratios, if you like) yields smoother bass. As I said earlier, the science of this makes some sense. Whether Ethan chooses 20 Hz or 5% or 12.7 cubits for his "criterion" is largely moot - so long as he considers some sort of "criterion" as an alternative "solution." IMO.
And speaking of Bonello's "5% rule," someone else brought this up to me recently and I am at a loss. I have reread Bonello's paper three times and I am uncertain where this "5% rule" comes from. He mentions no such thing in his 1979 AES paper. What's your take on this? Is it in another paper I'm not familiar with? Is it an interpretation of something else he wrote? I'd love to get to the bottom of that one!
But, even if we assume Bonello did state such a thing, it would be applied to all the modes, not just the axials. Such was the subject of Bonello's studies. Hence, it would be a very different application than that of Gilford's "20 Hz rule."
Edited for typos - %'s and Hz's and cubits!
_________________ ---lovecow---
"You've got to be very careful if you don't know where you're going, because you might not get there." - Yogi Berra
Last edited by lovecow on Mon Mar 28, 2005 9:11 am; edited 1 time in total
eric_desart Recording Org Pro Audio Forums
Joined: May 23, 2003
Posts: 816
Location: Belgium Antwerp
With recording I mean any space that registers a signal via a signal in the room with the intent to replay it in other rooms. Whether control, tracking, mastering whatever.
A listening space is the audio end users space, where the signal is only meant to listen to. e.g; living, home theater whatever.
Lots of critical educated listeners, with a lot of background are very pleased by their listening environment, while those spaces shouldn't fulfil requirement in function of accuracy for recording.
One rarely will see a living room with huge bass traps even for people investing a lot in audio.
There our brain does amazing things.
_________________ Best Regards - Eric Desart
Also or Non Respected Past RO Acoustics & Design Moderator?
Paul Woodlock Recording Org Pro Audio Forums
Joined: Aug 7, 2001
Posts: 675
Location: Peterborough, UK
With recording I mean any space that registers a signal via a signal in the room with the intent to replay it in other rooms. Whether control, tracking, mastering whatever.
A listening space is the audio end users space, where the signal is only meant to listen to. e.g; living, home theater whatever.
Lots of critical educated listeners, with a lot of background are very pleased by their listening environment, while those spaces shouldn't fulfil requirement in function of accuracy for recording.
One rarely will see a living room with huge bass traps even for people investing a lot in audio.
There our brain does amazing things.
hehe
Yeah, smoke a couple of splifs.... Which will do the following....
Even out your frequency response
Improve stereo image
Improve detail
Improve the song
Give you the munchies
Which is just as well really, as you won't be bothered with making bass traps and reflection asborbers after toking
I'll have to think about your comments about 20hz. Seems odd. Sometimes that means that in about a week I'll have an epiphany of understanding. I thought my ear was built logarithmically.
Quote:
And speaking of Bonello's "5% rule," someone else brought this up to me recently and I am at a loss. I have reread Bonello's paper three times and I am uncertain where this "5% rule" comes from. He mentions no such thing in his 1979 AES paper. What's your take on this?
There are two things mentioned in the "Master Handbook Of Acoustics 4th" about Bonello (two references, whereas you mention you've read one reference three times).
The first is the next paragraph after the one about Gilford which I mostly quoted above, in a section called "Mode Spacing and Coloration". Perhaps I'm reading it wrong (it's happened before). I'll quote both paragraphs.
Quote:
How close together must adjacent modal frequencies be to avoid problems of coloration? Gilford [ref 4] states his opinon that an axial mode separated by more than 20 hz from the enxt axial mode will tend to be isolated acoustically. It will tend not to be excited through coupling due to overlapping skirts but will tend to act independantly. In this isolated state it can respond to a component of the signal near its own frequency and give its proportional resonant boost.
Another criterion for mode spacing has been suggested by Bonello [Refs 5 and 6] who considers all three types, not axial modes alone. He states that it is desireable to have all modal frequencies in a critical band at least 5% of their frequency apart. For example, one modal frequency at 20-hz and another at 21-hz would be barelyl acceptable. However a similar 1-hz spacing would not be acceptable at 40-hz (5% of 40hz is 2-hz). Thus we see that Gilford's concern was primarily how far apart axial modes must be spaced to avoid problems resulting from independant and uncoupled modal action. Bonello's concern has to do with separations to avoid degeneracy (coincident) effects.
If you'd like to post a few words on the last two sentences I'd appreciate it. I think it means that frequencies less than 5% apart are problems, and frequencies more than 20hz apart are unrelated (not a problem).
The second thing that's attributed to Bonello which I've seen several places is mentioned on page 348 and a graph on page 349 of MHoA-4th, is the rules that
Quote:
He divides the low end of the audible spectrum into bands 1/3 octave wide and consideres the number of modes in each band below 200hz. The 1/3 octave bands are chosen because they approximate the critical bands of the human ear.
To meet Bonello's [refs 5 and 6] criterion, each 1/3 octave should have more modes than the preceeding one, or at least the same number. Modal coincidences are not tolerated unless at least 5 modes are in that band.
Coincidences I believe are defined back on page 344 by Everest
Quote:
Zero spacings between modal frequencies are a common source of coloration. Zero spacing means that two modal frequencies are coincident (called a degeneracy by acousticians), and such degeneracies tend to overemphasize signal components at that frequency
Ref 4: Gilford, C.L.S. "The Acoustic Design of Talks Studios and Listening Rooms", Proc. Inst. Elect. Engs., 106, Part B, 27 (May 1959) 245-258. Reprinted in J. Audio Eng. Soc., 27, 1/2 (1979) 17-31
Ref 5: Bonello, O.J., "A New Computer-Aided Method for the Complete Acoustical Design of Broadcasting and Recording Studios", IEEE International Conference on Acoustics and Signal Processing, ICASSP 79, Washington 1979, p 326-329
Ref 6: Bonello, O.J., "A new Criterion for the Distribution of Normal Room Modes", J. Audio Eng. Soc. 29, 9 (Sept 1981) 597-606. Correction in 29, 12 (1981) 905.
With recording I mean any space that registers a signal via a signal in the room with the intent to replay it in other rooms. Whether control, tracking, mastering whatever.
A listening space is the audio end users space, where the signal is only meant to listen to. e.g; living, home theater whatever.
Oh Oh.
I've been using the words 'recording room' to be equivilent to the places where you'd put a microphone and an acoustic guitar and a flute and drums and where one does recording.
And 'listening room' to be equivilent to control room, mastering room, home theater, possibly a living room -- defined by the absense of a microphone where you do playback.
This discusiion of modes has got me thinking: since it is desirable in a small (< 2500 cu ft) listening room to use dimensions that are likely to distribute the modes well, i.e. evenly, wouldn't splaying at least one pair of walls actually help?
For example, between a pair of parallel walls 14 feet apart, you can predict 5 axial modes at five specific frequencies below 200 Hz beginning at about 38 Hz.
But between a pair of splayed walls that start 12 feet apart and end up 16 feet apart, what happens? I know it's more complex, but wouldn't you get many many more weaker modes (compared to the parallel walls) that are distributed infintely? Wouldn't you get an infinite number of axial modes beginning at 35 Hz (for the 16-ft distance) through 47 Hz (for the 12-ft distance) and multiples thereof? And isn't such a distribution infinitely better in the overall scheme of things for the purpose of hearing more accurately in the room?
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
For more information on advertising, investing ,
merging
or any other ideas you may have for this community" Feedback
Pro Audio forums, Pro Audio Dealer, audio reviews and all the moderating here is
volunteer. Please remember no-one is being paid to be here or deliver
hot coffee. Play Fair, be polite, patient and considerate to others. Title your
topics properly and do not slander anyone, ever online. Read this before your post here:Recording
Org Disclaimer
This site can be translated into 13 languages. 錄音工作室幫助下，新聞和信息，數位專業音頻論壇, Opname studio helpen, nieuws en informatie, digitale pro audio forums, Studio d'enregistrement à l'aide de nouvelles et d'information, forums de l'audio numérique pro, Tonstudio helfen, Nachrichten und Informationen, digitale Pro-Audio-Foren, Estudio de grabación ayuda, información y noticias, foros de audio digital profesional. help, pro tools, cubase, nuendo, DAW, AD/DA, microphone, preamp, compressor, equalizer, Music Education, Arranging, Composing, collaboration des musiciens,
professionelle Musiker und Ingenieure,
colaboración de músicos profesionales y los ingenieros lo que pensamos acerca de una banda
Our new beta site is now online for testing The New Recording.org CMS
A room with a better spread of room dimensions (ratios, if you like) yields smoother bass. As I said earlier, the science of this makes some sense. Whether Ethan chooses 20 Hz or 5% or 12.7 cubits for his "criterion" is largely moot - so long as he considers some sort of "criterion" as an alternative "solution." IMO. 
