1. I won't comment yet--I want to hear a few others first
2. Same as #1, with the exception of the fact that I don't have an answer.
3. Says nothing in terms of equipment
4. Same as #1, except I do have an answer
5. This is normal and what I really wanted to comment on. Only frequencies below 500 Hz are really effected by positon unless there is some significant non-uniformity in the room--say for example you have first point of reflection absorbed at one positon but not the other--then you would see a difference in the HF response. Otherwise, the higher frequencies shouldn't change significantly.
6. You really are a bunch of Cheese Heads. Go Hawkeyes!! (we play Wisconsin next week) |
C'mon Rives get off the fence...LOL |
I'm not responding until Rives posts his answers. Then i'll let him know if he is right : )
The rankings below are rated from best to worst in descending order.
Broadband frequency response linearity
1) C
2) A
3) B ( WAY behind the others )
Most solid, even & extended low frequency response ( hardest to achieve )
1) C
2) A
3) B ( WAY behind the others )
Highest average broadband spl*
1) C
2) A
3) B
Obviously, the one to go with is "C" as it offers the most linear frequency response, smoothest and most extended bass response and the highest broadband sensitity.
As to your other questions, i know nothing about everything : )
The one suggestion / comment that i would make is that you need to work on your room acoustics. My thoughts are that the first place to start would be with damping ( NOT diffracting ) the primary points of reflection. Since the frequencies that seem to be the most troublesome here are relatively high in nature, you don't really need anything that is real thick or offers extended low frequency coverage. You can probably get by with using some relatively thin "acoustic foam" that is properly placed. This should tame the "hot" upper mids and treble response that you are experiencing without really toying with any other part of the audible spectrum. This would improve linearity even further and help to level out the otherwise "hot" and potentially highly sibilant upper mids / lower treble that you are probably dealing with. Sean
>
* If we limit bandwidth from 40 Hz and up, "A" offers the highest average spl of the three. Even with bandwidth limiting though, "A" does not offer the solidity or evenness in low frequency response that "C" does. As such, "C" is still the clear winner even after trying to compensate for the lack of extension in "A" and "B". |
Hey Sean, your suggestions about dealing with 'audible' issues such as hot upper mids are consistant with what I'm hearing. I should have said that I'm using the 3.6's without any tweeter attenuation, and I think that has certainly raised the higher frequency output. So re-inserting the 1.2ohm resistor would probably tame the treble response somewhat, to the detriment of transparency, as seems to be the case.
I wondered if you had an opinion on which of the 3 locations is the 12, 17 and 20.5 location?....or is that an unfair question given the parameters listed?
Lastly, would you say that the combination of speaker and room could offer an even lower output in terms of absolute low frequency extension, or perhaps a higher spl on the 20, 25 and 31.5Hz readings. I've put quite a bit of time into extracting any kind of output at 25hz and below, including placement, some diffraction materials, custom, stands, marble supports under stands, and the Cardas x-over cable kit. Would you say that's about as low as they'll go, or do you think there is more room for noticeable low frequency output? That might be a question for Tireguy perhaps, but I would welcome any input.
Rooze |
Here's a hint, Go to the free simulator on our website: the Room Simulator . Now I don't even have to post my answers:) You will also note, that the position as Rooze found, has little or no affect on frequencies above 500 Hz. |
Have I missed something? I looked at your calculations and on the surface C sounds best but for what purpose I don't know. Your speakers could be placed in the center of the room immediatly adjacent to each other and actually sound (in audiophile terms) like crap. Why not give us the dimensions of your room and the physical location of the speakers so we can make meaningful comments. Re the bass - I think C is misleading - I think your low bass in C is nothing more than a room node and A may be in a bit of a suck out. PS I always fail tests. |
Rooze: I would not insert a resistor into the tweeter section until you address the primary points of reflection. If you do that and things are still too "hot" sounding, then go the resistor approach.
As far as bass goes, Newbee has a point. That is, you need to configure the speakers AND your seated listening position taking room nodes into account. Getting the speakers positioned is one thing but then throwing that away with a poor listening position is all too common. Anybody that thinks that you can't take the room into account and obtain optimum results is fooling themselves.
Having said that, diffusion / diffraction is typically only good at mid to higher frequencies. Bass is a completely different animal and far trickier to work with. I can't really make suggestions based on the information that i have here. The one suggestion that i would make would be to pick up some books on acoustics and read, read, read.
Newbee: Your comments about room nodes is kind of "there, but not there". Everything that you hear is a combo of directly radiated sound and reflections / room nodes. There is no way to avoid this so the smart people learn how to tame and / or incorporate them into the response that you hear at the seated listening position. It is impossible to achieve linear response at multiple positions spread over a wide area.
Rives: I can't agree with your comments regarding rooms not really affecting anything above appr 500 Hz or so. The measurements that Rooze provided support my beliefs and experiences. While a lot of this will vary from room to room and speaker to speaker, making such a "generic" statement put you out on a limb. Luckily, i have a saw and a ladder and i'm going to get you down from there : )
Let's look at the variances above the cut-off frequency that you mentioned i.e. 500 Hz. Cutting you some slack, we'll double that and look at 1 KHz and above. I'll list the center frequency and the amount of variance from the highest reading to the lowest reading. This will demonstrate how much room loading can affect the sonic perspective at these frequencies.
1KHz - 13 dB's variance
1.25 KHZ - 9 dB's variance
1.6 KHz - 11 dB's
2 KHz - 6 dB's
2.5 KHz - 5 dB's
3.15 KHz - 5 dB's
4 KHz - 5 dB's
5 KHz - 10 dB's
6.3 KHz - 5 dB's
8 KHz - 10 dB's
10 KHz - 7 dB's
12.5 KHz - 6 dB's
As one can see, there are major divergences depending on room nodes / reflections regardless of frequency. To be fair to Rives, much of this has to do with the type of speakers being used. While Maggie's are not "oddball" or uncommon speakers, they do present a very different type of installation challenge than what most "front firing boxes" would. As such, one can not make sweeping generalizations without expecting some type of variance from installation to installation. Since i have every type of speaker known to man ( large towers, dipole's, omni's, mini-monitors and horns ), it is easy to see why i might have the outlook that i do whereas Rives has probably worked with more conventional designs most of the time.
As a side note, this is the reason that i don't respond to a lot of threads i.e. too much room for interpretation due to variables and not enough information presented. On top of that, i hope that some folks realize that "spec's" can be interpreted usefully. That is, IF one knows how to understand the data provided and that data was obtained in an accurate manner.
Having said that, i worked with the information presented here. I took into into consideration that the speakers are where they are and that Rooze has ( hopefully ) done his best to get them in the right spot. If such is not the case, then speaker placement should be done first and then one can calculate proper placement for room treatments from there. Not all rooms should be treated identically or following set formulas as the radiation pattern of the speakers to be used will alter what is required. As such, you can set up a room for one set of speakers and end up requiring a different configuration when you change speakers. Sean
> |
Rives, thanks, I've seen your site and tried the simulator, but I can't get my room dimensions keyed in, as the max width on your simulator is 10m and my room is 15m wide.
Newbee....my room is 45'X 28'. The speakers are 12' apart, 7.5' from the front wall and toed in slightly. I have some restrictions on where the speakers can go. They can't go closer than 18" to the front wall, and can come out into the room past 9' due to the proximity of a staircase. The closest possible listening position is at the 12' point, and the 20.5' position represents the farthest distance, being up against the back wall. The seat can go anywhere between the 12 and 20.5'....Though 12 is a little close due to the staircase and WAF.
There is an 'unreal' change in sound pressure as I walk between the 12 and 20'. At the 16-17' point, the spl just drops way down, and the low frequencies almost dissapear, then it just pops right back in there around the 19-20' area. I'm not really familiar with the term suck-out, though I guess that's what I'm experiencing in the 16-17' area.
To give some more information....the points ABC on the chart are actually A=12' B=17' and C=20.5'
I was amazed that the SPL could be significantly higher at 20.5' than at 17'.
As Sean stated, the best sonics would appear to be at the 20.5 position, given the apparent smoother low frequency extension. In practice, the sound is much better at point C (20') than at point B (17'). But at point A(12') the presentation is completely different than that at point C, even though the frequency/spl readings are quite similar.
There are obviously points in between the 3 positions that are possible, but only really at +/- 12" or so from the 12/17/20.5 points.
Anyway, it's been an interesting experiment for me. Right now I have the seat at the 20.5 position. I'm enjoying the
presentation of orchestral music and some big band stuff.
When I play more 'intimate' music like Diana Krall Live in Paris, or the Sting Live CD, I wish I were back at the 12' point for that enveloping soundfield effect!!...maybe I should install tram-tracks or something.
Anyway, when I moved to this bigger place and splurged on new 'bigger' sounding equipment, I never knew there was going to be so much involved. I wish I could afford to pay for the services of a pro installer who could perhaps get me at the 14-15' point by using acoustic treatments.
What do you think Rives?....I'm 12 miles from the Packer Stadium...I could get you tickets to watch some real Football, and you could fix up my room!...seems like a fair trade!
Rooze
|
Sean, sorry I just posted my last response as you were writing yours, so didn't see/comment on your last observations. What you say makes good sense, particularly that most of the issues concern higher frequency brightness and sibilance effects.
I'm intrigued to understand more as to what is causing the drop in SPL at the midway (17') point. Could it be that the 17' point is actually the more neutral place and therefore the 'best' place sonically?....perhaps what I'm hearing at the 12 and 20' points is 'boom' and 'refelections' for want of better terms. The sound is good at the 17 point, but just isn't warm enough with the absence of the lower frequencies...
Believe me when I say that as far as trial and error goes, I've worked bloody hard sliding those blasted marble slabs around!....I think where the speakers are now gives me the best low frequency extension and depending on the final choice of listening position, either a wide open soundstage (far position), or an envelopling, detailed and intimate sounding presentation (near Position).
Thanks again for helping me understand what is going on and help in exploring my options.
Rooze |
Rooze, I'll have to admit that I have never worked with a room this size, let alone a room of this size with panel speakers. For what its worth I would opt for the best positioning for resolution and soundstage and supplement the bottom end with a fast pair of subs located in an area which will give you flat bass to 20 hz. How's that for a cop out! By the way have you checked out other room set up systems, such as Cardas. If not, you can read about this and others in the FAQ's at the Audio Asylum site. I'm not sure how you came by your listening seat positioning, but did you try the 14 to 15 ft area. I've often found some of my best bass reinforcement when the speakers and the listening position are around 25% of the room dimension (which in your case would be 7ft for speakers and seat). Have fun - good tuning takes a long time and you should take your time do it. Its a lot more than just good measured frequency response. |
While i didn't present all of the figures involved in coming up with the conclusions that i did, C and A are quite close when looking at the big picture with B being off in left field ( no offense to left fielders : ) If going strictly by the figures though, C is the winner. Then again, facts & figures never factor in personal preference.
Newbee brings up some VERY valid points regarding low frequency response and output levels. With a room this large, you really should be looking at an active crossover and some subs. I would only go this route AFTER you feel comfortable that you've gotten everything dialed in with what you already have. Otherwise, you'll be running back and forth between the mains and the subs and it will only get more confusing. Sean
> |
Sean:
I didn't say rooms don't have an effect above 500 Hz--that would be REALLY wrong. I did say that positioning (both listener and speakers) has little effect above 500 Hz. Now that's not true in all rooms, but a room that is reasonably well balanced it is true. If high frequencies are tipped up, that will be true throughout most of the room--unless you are right up against one area that is heavily absorbing, but then this wouldn't be a well balanced room. In the bass region however there are peaks and nulls and they vary greatly by location. |
Rives: I'm sorry if i misinterpreted your comments, but i still don't agree with your clarification presented here. Due to the fact that all drivers "beam" or alter their radiation pattern as frequency is raised, one is bound to encounter varying frequency responses as speaker position and / or seated listening positions are altered. This can easily be measured outdoors where there are "minimal" room boundaries to cancel / reinforce / reflect the signals being produced and measured.
As further evidence, this can also be seen in just about any "decent" speaker review as the frequency response is altered as one changes the axis that one is listening on. Since distance changes the listening axis ( speakers don't rotate to accomodate seating distance ), you are bound to have quantifiable* differences in frequency response linearity. Sean
>
* don't know if this is the most appropriate terminology, but it sure sounds impressive : ) |
Sean: You are right about being off axis, and I was not very clear in my comments--since we were talking about a distance from the speaker the angle of incidence doesn't change much, a highly directional speaker could have a modest difference. Now if you had a well absorbed boundary at one point (first reflection) and a highly reflected surface at another it would yield quantifiably different results, but otherwise they will be remarkably similar. Try it, Rooze did, and I'll bet you get the same or very similar results. Keep in mind--I measure a LOT of rooms under many conditions, and my dealers measure even more than I do and send in the results. I have measurements taken at a variety of locations--I do not do averaging (it masks real data). |
Rives: I hope that you don't take this as a personal attack, only an open discussion of differing points of view.
I assume that Rooze basically took these readings walking away from the speakers in a straight line in what would be termed the "sweet spot". Then again, this is not mentioned but would be taken for granted as being the common sense approach by most readers.
As a side note, ALL speakers are directional and beam sound as frequency is raised. Obviously, some are worse than others in this regard, but they all do it. This happens not only horizontally, but also vertically. On top of this, the angle of incidence can vary quite a bit as distance is altered. Much of this will depend on whether the speakers are flat-faced or are toe'd ( sp ??? ) in.
If you look at any frequency response chart of a driver, off axis response becomes FAR less linear as one is moved further off axis. Going from 10* off axis to 30* off axis, which isn't that much, can make a huge difference in terms of frequency response linearity. Not only will this affect tonal balance, but also the soundstage, imaging and transient presentation of the system.
I have no idea what you or various dealers are using as a reference in terms of measuring frequency response and / or dispersion patterns. I would hope that it would be measurably more accurate than a stock Rat Shack meter. The non-linearities of this device itself, primarily in terms of frequency response and directionality, make the results very "questionable" under anything less than optimum conditions in skilled hands. As a generic tool used in the hands of civilians, it can be quite useful to study trends and average out results. As such, averaging is not a bad thing so long as the results are viewed in the right perspective. After all, Rooze verified that my observations, which were derived from averaging and following trends, were pretty consistent with what he was hearing.
Suffice it to say that we will probably end up agreeing to disagree here. Sean
> |
Sean/Rives, I had a question, the answer to which might help demonstrate each of your points of view.
I was pretty astounded at the effect of simply walking backwards through the room, from the 12' point to the 20' point, centrally of course and facing the speakers (which are now toed in only around 5* since the listening seat is so far back).
Forgetting about mid to heigher frequency anomolies and low frequency response, the thing that got to me was the shear change in volume... moving back through the 16-17 area, the volume reduced considerably then increased sharpley at the 18-19' area. I know this is somewhat indicated by the spl reading on the chart, but the actual experience of hearing this in the room was quite surprising and almost unnatural!!. Having had my original seat somewhere in the 16-17' area for a while, I'd been fighting gain issues with my LS15 preamp (another thread posted a while ago). Even after installing the ARC approved hi-gain modification, I couldn't get satisfactory volume in the 16-17' area. Now I'm sitting some 3-4' further away from the speakers, the volume has increased to more realistic levels without further equipment changes.
I'm curious to know more about the 'theory' of what is happening here. Is the 'dead-spot' at the 16' area actually the most neutral place to be? As mentioned previously, the sonic's lacked some depth and warmth and certainly lacked absolute low frequency extension in the 15' area.
Is this spl phenomenon common in room installations of these dimensions?
Is there a way somehow to treat the room to create more spl in the 15-16' area?.....this I would think is the most desireble area to sit. I've heard others offer as a rule of thumb:- measure the distance between speakers (12') and add 2-3 feet for the distance to the chair (15').
So I'm curious as to what changes in the room might equalize the spl more in the 15' area with that which is heard at both the 12 and 20' points.
Given that there is so much clearance between each speaker and the side walls, is it impossible to introduce any kind of side-wall reflective, diffuser or absorber that could be angled (perhaps) to create a higher spl in a given zone. ( the room width is 45' with the right side speaker approx. 10' from the side wall).
I've always understood the importance of room acoustics in arriving at good sound, but I've never had this kind of experience where differences are so marked within different 'zones' of the listening space.
Rooze |
This is strictly a guess, so take it for what it is worth.
My "guess" is that the reflection from the back wave of your Maggies is "colliding" with the front wave at these distances. As such, you're experiencing out of phase cancellation, reducing the spl. The reason that this takes place over a specific area is that sound waves are all different lengths. One batch of frequencies is nulled at 16', another batch at 18' and the frequencies between those points at 17'. As such, you have a noticeable broad band dip at this specific distance.
In front and behind those distances, you'll experience reinforcement of those frequencies ( to varying degrees ) but cancellation of other frequencies. The key is to find the point that allows the most neutral tonal balance while still allowing good to optimum soundstaging / imaging. Since a microphone and test gear don't hear or process information the same as our brain does, using equipment and tools to get you in the area is fine, but trust your ears.
As far as doing the math to figure out the wavelengths and reflected paths, a room with irregular shapes, sizes and / or irregular non-solid objects that break up reflections can cause pretty erratic results. This is the reason that the acoustic modeling program called C.A.R.A. wants to know as much as possible about your room, it's shape and size, what you have in it, the speakers radiation pattern, etc... The more info that one can provide an acoustician or a program similar to CARA with, the more likely the results are to be accurate.
As far as "boosting" the spl's and linearizing the frequency response at that distance, this can be done. It will take a lot of work and will completely change the presentation that you hear at any given distance. This is because you'll be "squashing" the back wave to minimize cancellations and reflections. In effect, you will probably end up losing many of the desirable attributes of the Maggies that attracted you to them to begin with.
As a side note, with Maggies at 12' apart with a very mild amount of toe-in, 16' - 18' is too far back in my opinion. Then again, i don't know the specifics of the room or anything else about this installation, so keep experimenting and learning as you go. This can be a very educational and beneficial time for you in terms of learning via first hand experience. If you really get serious and start doing a lot of experimenting, keep a notebook. I would also make cohesive notes, not just jot down things at random. You might not look at these notes for a while and by that time, you may have forgotten some of the specifics of your "abbreviated" notes. Sean
> |
Sean, you make the point that speakers beam at higher frequencies and are less linear off axis. Are you then of the opinion that speakers should be toe'd in for best overall frequency response?
If there is a high frequency hole at the listening area, can it be boosted by toeing in the speakers to get a more direct on axis presentation?
I am aware that while many speaker placement methods recommend some toe in, at least one major speaker manufacturer states that best results are achieved with little or no toe in.
Also for both of you, what are the effects on sound stage width and depth as one adds toe in? Can it be predicted or is it room dependent?
Thanks, |
I did a post not that long ago giving a brief run-down on the method that i generally use for setting up speakers. It is somewhat of a universal approach that works regardless of room or speaker size and takes into account tonal balance and soundstage / imaging. Only problem is, i can't remember what thread it was in and it's too long to type out again. I'll see if i can find it or hopefully, someone else will point us to it. Sean
>
PS... As a general rule, i prefer little to no toe in. Then again, that is with using the approach that i mentioned above. |
Aargon: Found the thread that i posted a basic and relatively universal speaker set up method in. It it just so happens that Rooze started that one too : )
Look for a thread of his entitled "problems with Maggie 3.6's" or something to that effect. This should answer your questions and then some. Sean
> |
Thanks, your method is straightforward. I put my 5As on baking pans and started pushing them around per the process earlier this evening. It was easy to hear the modal variations, and detect when the female voice image was solid. I will back this up with measurements from my RTA in the AM. |
