RFZ_Quest - Quite a listening room. A bit larger than before. I'm curious about the floor treatments under and in front of the speakers. What effect does having mats and foam underlayment have on the sound quality? I assume the mats are resonance related. As to the foam -- is it a purely theoretical action or do you notice a difference? I ask as I'm completing my room and have wondered about the effects of having a framed wood floor.

Paul,

Yet again I should have read Steve's white papers before asking a question. Thanks for the reply! It seems that room treatments are somewhat analogous to simultaneous and multi-dimensional games of pool, with lower freqs tending to have quite a bit of spin.

After checking out the comments on the new ERR's, and ones in this forum, I've decided to permenently locate the RL3's in my own great room (instead of the little upstairs bedroom as planned). Will be making full use of diffusors and absorbers  -- fortunately, the speakers are in front of an easily treatable high wall and the room's sides can be softened by letting down the fabric woven window shades, effectively creating 6-inch deep diffusors (if I'm not mistaken). Still mulling over the rear fireplace situation though (a huge brick and metal structure about 20' from the speaker fronts) ... that'll be a challenge, but no problems in the 'wife friendly' arena at this time ... we're both just enjoying the sound!

Keep us posted! Pics are great - really helps. Room treatments need more press.

Tim

Wolfgang,

Not sure what the "diffusion-speaker" concept involves, but I have serious doubts as to it's general feasibility for this purpose.

If it is what I imagine it to be, then at best I would speculate very slim diffusion factors when compared to properly designed diffusion modules.

Of course, any diversion from a flat wall surface is going to provide some degree of diffusion. Actually, the effect uncontrolled is more like that of random reflection rather than that of true diffusion. It is usually in the form of non-unity in it's spread pattern and reduction of delay.  Offset, or canted wall techniques, book shelves, furniture, etc., offer only a limited degree of enhancement when the factors of diffusion coefficients come into play.

With predetermined factors based upon mathmatic calculations for which to center precisely upon a desired coefficient rating, we know with close proximity what to expect in overall performance for a well modeled design.

Not all diffusion concepts or designs are created equal. There are many variances which separate the performance rating from one design to another.

For instance, some designs are based upon a specific limited range as that is the area of concern for it's use. For a broader bandwidth and greater coefficient factor, designs require much larger and far more complicated  design parameters.  

Many do not understand the complexities of how and why diffusion principles work. It is important to understand this in order to build and utilize these units in order to have any sort of predictability as to the outcome of their use.

If one was to build a unit corresponding to 'Half-Wave' measurements instead of the required Odd-Quarter-Wave parameters, the unit would virtually be rendered useless. The unit simply would not work, as the timing would nullify the effect desired in the sequencing factor.

A good diffusion sequence will provide an even spread in time over a large area as opposed to the original wavelength of the incident sound impinging upon it.

The diffusion principal is based upon number theory sequencing in specific order and of specific design. The calculations are always prime numbers arranged in sequence of ODD-ORDER-Quarter-Wave function in order to work effectively.

Just merely creating random surface textures offers very little benefit by comparison.

Remember, if the surface of the diffusion module is not hard enough to block the sound from penetrating it, it would be unlikely that the sound could be redirected properly within a diffused pattern back into the room. Otherwise, the result becomes more of a refraction or even an absorption effect with totally different results than what was expected.

There are so many variables to consider when designing diffusion modules. It is very easy to build them incorrectly with less than desirable results when parameters are not in accordance with the math that dictates their potential.

Hope that gives you some sort of food for thought concerning this issue.

Paul.

The RPG nested grouping in this diffuser is an excellent approach.  It is however complex to build.  It is based on the same quadratic theory as all prime 7 sequence diffusers, meaning it is solidly based on math.

Steve

I believe they have the plans available for the RPG on their web site.  In theory the well dividers should be as thin as possible.  

Steve

Hello Wolfgang,

The unit in question presented by RPG Systems is a complex integrated design, which is basically a diffuser within a diffuser concept. This means that multiple stages of diffusion is built-in together to form a very wide coverage of the audio spectrum.

The first stage consists of a lower prime designator, which sets the stage as  'building blocks' for the more intricate second-stage sequence within each of the initial lower prime wells.

Each of the large prime-7 wells actually becomes a discreet diffusion module within it's own boundary.

The benefit gained from this approach is that a much wider degree of diffusion coefficient can be realized in a very effective manner. It is simply a means of integrating multiple diffusion elements into one high performance model, which is truly wide-band in coverage.

BTW: All RPG designs are based upon highly complex number theory sequences. The people behind these designs are the most respected in the industry. I have yet to find any of their products as anything less than ideal!

There are calculators available for this purpose with the Modulo function in memory. With this function, number theory sequencing becomes a simple procedure based upon the prime number setting of choice. Steve provided an earlier example of this when he displayed a variation of my design, broken down into the exact mathematical equivalent for which to duplicate within close proximity.

Just follow the numbers and connect the dots. The end result will reflect the degree of accuracy in the design build.

What RPG did is rather simple in design although it becomes more of a challenge in construction.

Something like this concept usually requires a large wall based integrated system due to the layout required. The outer shell is based upon 'lower' Prime number theory, with more complex prime configurations built within each of the primary wells.

If you want to build something like this, you have options at your disposal, which allow for a wide degree of flexibility in overall design parameters.

A system like this is like building a complex crossover network for a 3-way driver speaker design. You are actually integrating multiple 'crossover points in unison to cover the largest range of the frequency spectrum as possible in an effective manner.

I have strongly desired this type of concept for quite some time. This level of build is reserved for the full-scale approach. This is when the walls themselves become the building platforms for which the transformation takes place AS the diffusion element, rather than a backdrop for mere enhancement placed in front of it.

The general rule for diffusion coefficient performance is this: The 'deeper', and the 'narrower' of the well layout, the higher degree of bandwidth coverage possible. The higher the Prime number, the higher this layout grows in significance.

The parts layout, which went into my Quad Prime-23 unit, is staggering in size, weight and cost, due to its complexity. It is however, worth every bit of its trouble in building as the performance quality is equally staggering by the impression it relays. My unit maintains a complex blend of absorption as well as diffusion principles, incorporated effectively into one large-scale design.

If you want the design parameters to replicate something along RPG's design, I can easily formulate a respectable clone with the flexibility of options to choose from.

To keep things on a practical perspective, the basic design parameters must be considered and adhered to.

Space and size is always going to be the first and foremost thing to consider. The overall effect and degree of performance obtained within a practical platform will then become the second factor for which to base your design build.

You must first decide the exact level of performance required meeting your goal, and whether or not this level of performance can be utilized within your space requirements. As you can imagine, a full sequence of wells built within each of the primary wells automatically dictates a considerable amount of real estate. It is however, a fantastic design principal with very efficient results. I would highly recommend this concept if you are willing to build it correctly.

I suggest a maximum configuration of a Prime-7 base platform, with a Prime-13 sequence integrated within each of its wells.

If you wanted to build a massive unit encompassing the entire size of your wall, this is where the unit advance's to a triple stage, extreme bandwidth layout where shear size is necessary to make it possible.

This means that you have actually placed a full prime number sequence 'within' an individual well of another sequence, again WITHIN another individual well of the primary base sequence.

How's that for coverage on a grand scale!

BTW: ALL dimensions are critical in respective to the outcome of the units performance. Whatever the parameters are shown as the result of the math is what you MUST adhere to or this all becomes an expensive waste of time with very little relevance to it's outcome.

I cannot stress this enough! You either have a unit that does what it is designed to do by strict adhesion to accuracy, OR a unit, which simply LOOKS the part but does very little to nothing by comparison.

For maximum effectiveness, it is imperative that full divider panels are utilized between the wells when a quadratic-residue design is considered. You could eliminate the use of divider panels with 'Primitive-Root' Sequencing. This will never perform on the level, which the more sophisticated quadratic principle provides.

Sorry, there are NO shortcuts here if you want maximum performance. The rules are dictated by the laws of Physics. It is as simple as that!

Follow the golden rule precisely and you will be rewarded for it in the end. Ignore this fact and you will return to where you started from.

So, the question is this;

Where are you willing to go with this layout and what is your ultimate objective?

Give that a good deal of consideration, then focus toward accomplishing your goal.

I believe that I can lend the information you seek to accomplish this task. This I can provide as time allows.

Never be afraid to push the envelope of possibility as long as knowledge and dedication remain on your side!

Paul.

Wolfgang,

I'll try to answer this as easily and to the point as possible.

The Fibonacci number system ironically holds a very strong correlation to just about all natural factors in life, thus of which I'm sure maintains a bit of relevance towards the harmonic qualities of music within itself.

However, I think there is a bit of confusion here toward the principal in question and how that it affects the outcome of QRD parameters.

The Fibonacci system-starting sequence begins with zero followed by a one. Each subsequent number is equal to the sum of the previous two numbers of the sequence itself. I.E. After two starting values, each number is the sum of the two preceding numbers.


Our system of concern for determining quadratic residue diffusion sequences is based upon prime numbers exclusively. The prime number (modulo N) determines the length of the periodic sequence (the number of wells that the sequence consists of).

The greater the number of wells (per increasing prime value), the greater the frequency range in which the diffuser is effective.

The deeper the wells are, the lower the extension toward the lower limiting frequency at which sound can still be diffused effectively. The depth is determined somewhere around one-half the wavelength of the lower limiting frequency.

The well width determines the extension of the higher frequency cutoff. This is calculated at a 'half-wavelength' based on the shortest wavelength to be scattered. Whatever the highest frequency you desire to base your workable diffusion coefficients upon, thus becomes the shortest wavelength for which to determine your well width.

If you desire an extension of 20K, then you simply divide the speed of sound in air as the medium (1,130 ft per second) by the frequency to determine the wavelength. In this case we are looking at a figure of around 0.05 in which to base the well width on.

These prime numbers are 2,3,5,7,11,13,17,19,23,29,31,37,41,43,47,53,59,61, etc....

With these, we have the basis for determining a feasible unit for which to meet our objective within the periodic, quadratic type of diffusing system.

The well depth sequence is determined from the quadratic residue sequence, based upon a prime number, N. This is known as the well depth proportionality factor which = n2 MODULO p.

p= a prime number & n= a whole number between zero and infinity.

Modulo simply refers to residue. This can also be related to the expression: "the remainder after dividing by".

Example: If I wanted to design a QRD based upon the prime # 29, I would come up with the following sequence for which to determine my well depths.

well 0=0" depth.
      1= 1"
      2=4"
      3=9"
      4=16"
      5=25"
      6=7"
      7=20"
      8=6"
      9=23"
      10=13"
      11=5"
      12=28"
      13=24"
      14=22"
      15=22"
      16=24"
      17=28"
      18=5"
      19=13"
      20=23"
      21=6"
      22=20"
      23=7"
      24=25"
      25=16"
      26=9"
      27=4"
      28=1"
      29=0"

Always start with a 'zero-depth' well factor at the beginning of the sequence. Your final well in the sequence will always terminate as a zero-depth well, which means that there is actually one more well than the prime number it is factored upon. This is due to the fact that the sequence starts with zero and not one.

When two periods of this sequence are repeated, it is only necessary to have a single 'zero-depth' well between the two periods. It is also desirable to design the zero-depth wells at half-width of the main wells when possible.

All of these measures benefit in the reduction of periodic lobing, which is a major problem with QRD designs when continuos periods are fashioned. I will touch on this factor later as it is always of primary concern when designing the best possible deterrence towards lobing effects.

I will continue with this later. Don't worry about all of this technical detail. I just wanted to clear some things up about how these things are designed and what makes them tick.

We will get to the most important thing that only matters in the end. You will get the choice of building the units best suited for your purpose. The most difficult thing for you to decide on is the extent of methods used in the total scheme of things. Then you simply build it.

Hope that at least clears up some of the technical aspects in question.

Back with this soon.

Paul.