[andyr]

If you go here:
http://www.cardas.com/content.php?area=insights&content_id=41&pagestring=Listening+Room+Dimensions

... you can see George starts off with his "Golden Cuboid" room - where, for a length of 20', the ideal width is 12.36' (20/1.618) and the ideal height is 7.64' (12.36/1.618). That's all straightforward "Golden ratio" stuff.

Then he goes on to state ideal measurements for what he calls a "tapered trapagon" where the width of the back wall (behind the listening position) increases to 13.9', by means of one side wall being angled outwards slightly.

And the height has increased (which I'm not concerned with).

He then states ideal measurements for the optimum room - a "golden trapagon", where both sides angle out slightly so that the width of the back wall is 15.72' (whereas the front wall is 12.36').

I wish to understand how he arrives at 15.72, in relation to 12.36. The reason - we are designing a new house to live in and, like the one we've been in for 25 years, a "listening room" is part of the house specification. (Alas, quite a bit smaller than the room I have now.  )

Thanks if anyone can enlighten me. 

Andy

[JLM]

I did his "Golden Cuboid" design for my audio mancave when we built 6 years ago.  The ancient Greeks called this shape the golden rectangle and would be in proportions of 5 to 8 (very close to his numbers).  They did it because it looks good (not awkwardly tall or wide) but more importantly it cut down on the echos inside those stone (hard surface/no absorption) temples.

If you play around with the numbers you'll find his numbers (similar to what others recommend) don't easily work in multiples of each other (like 3 times the width equals 2 times the length), thereby reducing echo in one direction (say along width of room) adding with the those in another direction (say along length of room).

BTW shape is only 1 part of having a good room.  Isolation (physical/electrical) is another.  PM me.

[andyr]

I did his "Golden Cuboid" design for my audio mancave when we built 6 years ago.  The ancient Greeks called this shape the golden rectangle and would be in proportions of 5 to 8 (very close to his numbers).  They did it because it looks good (not awkwardly tall or wide) but more importantly it cut down on the echos inside those stone (hard surface/no absorption) temples.

If you play around with the numbers you'll find his numbers (similar to what others recommend) don't easily work in multiples of each other (like 3 times the width equals 2 times the length), thereby reducing echo in one direction (say along width of room) adding with the those in another direction (say along length of room).

BTW shape is only 1 part of having a good room.  Isolation (physical/electrical) is another.  PM me.

Thanks, JLM.

Yes, I'm all set with the straight golden cuboid requirements ... but, BTW, it is possible to have different combinations of L/W/H which give equally good room-mode results, according to the THX room-dimension calculator that I have.

So I was interested in exploring George's "trapagon" concept.  In fact, after my post, I did find the theory behind it - the area of the wall behind you should be 1.618 times the area of the front wall (where the speakers are).  You get this by increasing each of height and width by the square root of 1.618, which is 1.272.

But this would produce problems in my building - so I will stick with a cuboid room that I can optimise with the THX spreadsheet.

And in terms of electrical isolation - this new house will have 2-phase power in from the street, so 1 x 80a (240v) phase will be for the house and 1 x 80a phase will be dedicated to the music system.  IE. 4 x 32a circuits:
* 1 for digital sources (CD-R unit and Lingo PS for my LP12)
* 1 for analogue sources (phono stage, preamp and FM tuner), and
* 1 for each speaker active XO/power amps.

Regards,

Andy

[JLM]

Yes, I'm aware of other ratios, and as always bigger is better.

My 8x13x21 basement room includes:

Insulated fiberglass exterior door (with weather seals)
Lined flexible duct for supply air
Insulated staggered double stud walls

All contribute to making the room spooky isolated from the house

My downfall was the builder's refusal to suspend the drywall ceiling (using gauge metal furring strips) and use of recessed cans.  Even with insulation in the ceiling I hear the footsteps/etc. from above.  But another layer of drywall with green glue and swaping out the cans for track lights could fix that.

[ctviggen]

Thanks, JLM.

And in terms of electrical isolation - this new house will have 2-phase power in from the street, so 1 x 80a (240v) phase will be for the house and 1 x 80a phase will be dedicated to the music system.  IE. 4 x 32a circuits:
* 1 for digital sources (CD-R unit and Lingo PS for my LP12)
* 1 for analogue sources (phono stage, preamp and FM tuner), and
* 1 for each speaker active XO/power amps.

Regards,

Andy

You only have 80 amps for your entire house and 80 amps for your listening room?  Is that correct?  Are you using gas heat, gas hot water, gas stove/oven?  What about air conditioning?  I have 200 amp circuit for my house, and that includes my HT room, but I have an electric stove/oven, but oil furnace/hot water.  For my HT system, I have two single-phase circuits; each is 20 Amps, so 40 amps, although I have a lighting circuit (15A, single phase) and extra outlets along the walls for lights, etc. (20 amp, single phase).  However, in reality, my system draws nowhere near 40 amps.   

[andyr]

You only have 80 amps for your entire house and 80 amps for your listening room?  Is that correct?  Are you using gas heat, gas hot water, gas stove/oven?  What about air conditioning?

In Oz, a "normal" house has 1 x 80a feed from the street.  So that's what I will will most probably have ... with a second 80a feed for the hifi system.  (Sure that's overkill, in terms of amps used, but I'm doing this mainly for isolation ... plus it will allow me to have 4x32a circuits to power the system.)  For the "normal" part of the house, yes, it's gas heating, gas H/W, and gas cooktop ... but a 32a circuit for the oven.  No aircon.)

Regards,

Andy

[JLM]

Somehow my previous (older) house I'd bought had a rat's nest of a 60 amp service that had served all the typical stuff in an 1100 s.f. house (with finished basement) plus A/C plus well plus pump for a built-in pool.   

Our current house (twice as big) as underground service, our own transformer, all new appliances, all 12 gauge/20 amp circuits, and dedicated 20 amp circuit for each of my three cyro'd hospital grade duplex receptacles for my main audio rig (those 3 also have a separate ground).

[tubamark]

. . .  a "golden trapagon", where both sides angle out slightly so that the width of the back wall is 15.72' (whereas the front wall is 12.36').

I wish to understand how he arrives at 15.72, in relation to 12.36. The reason - we are designing a new house to live in and, like the one we've been in for 25 years, a "listening room" is part of the house specification. (Alas, quite a bit smaller than the room I have now.  )
Andy

AREA! :

front wall 94.43 sf
rear wall 152.484

front/rear =  0.619

These ratios aren't really as critical as Cardas makes it out to be, however.  Many other intervening variables exist in the real world that change the effective geometry of the room.  Given an existing cuboid room of the same volume to work with, I would probably stick with the shoebox, rather than sacrifice room volume.  A front wall 7' x 12' or smaller is pushing-it, unless one is planning for only one or two good listening seats.

Having said that, if I had the luxury of building a room entirely from scratch and wanted a moderately live room, I would choose some variation of taper.  In addition to reducing echoes without absorption, etc, it also shifts the first reflection points (from sidewall and ceiling) rearward, possibly to a point behind the listener.  The first floor reflection can be easily addressed several different ways (No, carpet ain't one of 'em). 

-- Mark

[oldmp3]

A very interesting article (http://www.acoustics.salford.ac.uk/acoustics_info/room_sizing/?content=index) about computer modeling for:

"...producing the flattest possible modal frequency response for the room to determine the best room size. It uses a computer algorithm to search for best solutions [5]. Furthermore, the algorithm has been further advanced since reference [5] to look at room dimensions which are robust to changes in room size due to construction tolerances in terms of the room size and the properties of the construction material..."

They looked at room volumes of 50, 100, and 200 cubic meters, and provide ratios that are found to be scalable across reasonable room sizes.  Interestingly, the "best" ratios cluster around 1:2.19:3, and they provide some "second-best" ratios.

One of their main points is that "fixed" ratios like the "Golden Cuboid" (1:1.6:2.6) don't necessarily scale for different room volumes. And it doesn't come close to their "best" or "second best" values.

I used their ratios to model some room sizes for a planned listening room with a fixed ceiling height and constraints on length and width.

[tubamark]

. . . One of their main points is that "fixed" ratios like the "Golden Cuboid" (1:1.6:2.6) don't necessarily scale for different room volumes. And it doesn't come close to their "best" or "second best" values.

That's right!  I don't want to beat-up on Cardas, but all one has to do is plug-in the dimensions of a golden-ratio rectangular room into any modal calculator (many free ones out there, or diy in Excel),  and you'll see that it actually produces some duplicate resonances, more than many other ratios out there.

Having a gently tapered room is a good idea, but doesn't actually make a big difference in terms of modes - which is the main argument often cited to favor the practice.

The problem is that so few bona-fide studies have been done comparing all the modeled responses to real rooms.  Real rooms tested rarely behave as predicted by a simple dimensions-only calculation, and only begin to behave as predicted when using a concrete bunker with no windows or doors.

Room ratios are worth considering, but not to the degree that many claim.  A room with perfect dimensions still has as many modes as the next one, and speaker and listener placement determine the outcome as much as the dimensions themselves.

-- Mark

[Ethan Winer]

all one has to do is plug-in the dimensions of a golden-ratio rectangular room into any modal calculator (many free ones out there, or diy in Excel),  and you'll see that it actually produces some duplicate resonances, more than many other ratios out there.

Indeed, and all the other "recommended" ratios also have uneven mode spacing. In the grand scheme of things, there's no such thing as a perfect ratio. And even if there were, the modes still have ringing peaks that need to be damped with absorption. And of course nulls that also need bass traps to improve.

--Ethan

[oldmp3]

A very interesting article (http://www.acoustics.salford.ac.uk/acoustics_info/room_sizing/?content=index) about computer modeling for

Ethan, can you speak to any of the technical details of the article or paper cited (requires AES membership or purchase)?

I tend to think there probably are a set of ratios for cuboid rooms which are at least good starting points, from which you can depart to treat peaks and nulls, or whatever other artifacts of the room you consider undesirable. Their aim was to predict ratios with "flattest possible" modal response in theoretical rooms.  Wouldn't these be a good starting point?

[Mitsuman]

I wish this man had published his thoughts on the subject as it pertained to home listening rooms.

http://acousticalsociety.org/about/awards/sabine/12_10_10_harris

All I know is he did a damn fine job with Powell Symphony Hall here in St. Louis. (amongst a few others.  )

http://www.stlsymphony.org/media/pdf/charts/orch1112.pdf

[Ethan Winer]

Ethan, can you speak to any of the technical details of the article or paper cited (requires AES membership or purchase)?

I don't know which paper that article refers to, but I'm quite familiar with the concepts.

I tend to think there probably are a set of ratios for cuboid rooms which are at least good starting points

Sure. But the notion that there's a specific ratio that spreads the modal peaks out uniformly is incorrect. Some ratios are definitely better than others, but none are perfect. And as I mentioned, even with a very good ratio - with modes spaced as uniform as possible at least for the lowest octaves - bass traps are still needed to reduce the inevitable peaks and nulls and ringing.

--Ethan

[Mitsuman]

AREA! :

front wall 94.43 sf
rear wall 152.484

front/rear =  0.619

These ratios aren't really as critical as Cardas makes it out to be, however.  Many other intervening variables exist in the real world that change the effective geometry of the room.  Given an existing cuboid room of the same volume to work with, I would probably stick with the shoebox, rather than sacrifice room volume.  A front wall 7' x 12' or smaller is pushing-it, unless one is planning for only one or two good listening seats.

Having said that, if I had the luxury of building a room entirely from scratch and wanted a moderately live room, I would choose some variation of taper.  In addition to reducing echoes without absorption, etc, it also shifts the first reflection points (from sidewall and ceiling) rearward, possibly to a point behind the listener.  The first floor reflection can be easily addressed several different ways (No, carpet ain't one of 'em). 

-- Mark

And volume. 

http://accessscience.com/content/Reverberation/584900

[FullRangeMan]

Dimensions: Front: 3.00metres x 4,85metres   Rear: 3.78metres x 6,00metres   Long: 7.85 metres

[FullRangeMan]

Also from Cardas Site: monopole speaker position in a Golden Cuboid room, according the Golden Rule.

[andyr]

Also from Cardas Site: monopole speaker position in a Golden Cuboid room, according the Golden Rule.

Sadly, he doesn't give a similar suggestion for dipoles (Maggies)! 

Regards,

Andy

[Rocket_Ronny]

What I have found to be most important besides room size ratios is simply Room SIZE. Too small a room sound closed in while a larger room like my favorite 26'x16' or 18"x10 is very nice and open. This gives the speakers enough space to breath. As long as you don't make the mistake of having them by a wall. I like mine out 12'.

Rocket_Ronny

[FullRangeMan]

Sadly, he doesn't give a similar suggestion for dipoles (Maggies)! 

Regards,

Andy

The site no more had the Dipole Calculator??  Afew years ago I made the calcs at the site for Dipoles, that are this image below, hope this help: