[whydontumarryit]

Hi James,

How can I get a flat frequency response within +/-1 db between 60hz and 10khz at the listening position with a pair of middle t speakers without using equalization? How close to that are the speakers in one of your rooms?

Thanks

[James Tanner]

Hi James,

How can I get a flat frequency response within +/-1 db between 60hz and 10khz at the listening position with a pair of middle t speakers without using equalization? How close to that are the speakers in one of your rooms?

Thanks

Hi,

I would say in the real world plus or minus 1dB is impossible with any speaker in a listening room.

Using EQ will totally screw up the Sound Power.

Middle T - Anechoic measurements:




james

[GrooveControl]

How can I get a flat frequency response within +/-1 db between 60hz and 10khz 

You can waste a lot of time and money trying to achieve that, only to discover you don't like it.  Especially at low volumes.  For me, the following response chart is very pleasing. Hardly +/- 1db.  YMMV. 

@ James, your chart looks incredibly flat.  Just curious, how do you define the "Listening Window"? Is it on axis, or an average over +/- xx degrees? 

EDIT: @ James, please discard.  I found a definition of the term.  Simply put, an average of 5 measurements taken - on axis, and 15 degrees left, right, above and below axis. 

BTW, my measurement is using the 'moving mic' method at my listening chair.

[James Tanner]

Hi Groove - yes the 5 measurements are the way most do it but we actually do 360 degrees around the speaker both Vertically and Horizontally for the Sound Power and Listening Window in the anechoic chamber (over 300 measurements).

Its what we call our "Family of Curves" I am doing a Newsletter on that now.

james

By the way you need a large Anechoic Chamber in order to make these types of measurements - which we have in our factory.

[whydontumarryit]

You can waste a lot of time and money trying to achieve that, only to discover you don't like it.  Especially at low volumes.  For me, the following response chart is very pleasing. Hardly +/- 1db.  YMMV. 

@ James, your chart looks incredibly flat.  Just curious, how do you define the "Listening Window"? Is it on axis, or an average over +/- xx degrees? 

EDIT: @ James, please discard.  I found a definition of the term.  Simply put, an average of 5 measurements taken - on axis, and 15 degrees left, right, above and below axis. 

BTW, my measurement is using the 'moving mic' method at my listening chair.

That is a fine looking in-room response you have there Groovecontrol.
Close enough to within 2db to suit anyone. Although, obviously not possible unless REW or the like is used to flatten the low end.

As you can see from the photos of James' listening rooms it appears that nothing more than front wall treatments are used to tame the room and he has made it clear his views about EQ.

The problem is while still being the best speakers I have ever heard, reading the glowing reviews from people hearing James' systems (even the Mini A!) makes me think something is missing and that he would be able to shed some light on this by describing his methods of speaker location, room setup, etc. unless the information (link?) has already been discussed.

I asked if he had personal in-room measurements not marketing copy. Much like you were kind enough to provide. For what speakers btw?

[James Tanner]

Hi Why

I just ran now a FFT response of my set-up at the listening location for you.



james

[GrooveControl]

You are correct, I'm using REW/miniDSP to equalize below 1Khz.  Here is chart showing the unequalized response in red.  Somewhat nasty.  Not sure if it's the room, the aftermarket recone kits, or most likely a combination of the two. Either way, the EQ helps a lot more than it hurts.  The speakers are JBL 4430. 

[James Tanner]

Hi Groove - take a measurement at 1 meter from the speaker on axis - that should tell you what the speaker is doing relative to the room.

james

[whydontumarryit]

Hi Groove,

Any so called audiophile younger than 40 yrs. would more than likely be disturbed that a pair of circa '85 buttcheeks would perform as well as what you show with the addition of modern technology.
The question is the audible difference between the choices. Commenting on a difference between my middle-t and a revel 226 or similar would be a waste because they are basically the same thing, designed for the same results based on Toole's work and hopefully more so on Geddes and Linkwitz.
Your choice of the JBL is a good example of a speaker that has everything going for it. Great dynamic range, controlled directivity, with not as wide a dispersion pattern as the middle-t (a difference I'm starting to question in terms of accuracy) and enough balls to eq in the LF response to get another octave out of them. Flat to 25hz, once heard, is something you can't give up and , in my opinion, a good percentage of why the middle-t sound so good and so engaging at a low volume

If you have attempted to 'update' your system with more modern loudspeakers and found it wanting I would be interested in an opinion of what brought you to your current choice and what the mini dsp and eq give and take away from the enjoyment.

thanks

[whydontumarryit]

Hi Why

I just ran now a FFT response of my set-up at the listening location for you.



james

Hi and thanks for the response.

Does a FR graph with 1/3 or 1/6 smoothing obscure what is heard.
 Is 1/12 or 1/24 showing high Q abberations that are not audible.
What should I pay attention to?

thanks

[James Tanner]

Hi Why

The problem with these types of measurements is it is simply a guide.  You are measuring a single pressure point in space. 

That is not what we hear - we hear the total radiated energy throughout the room - Sound Power.

So I use these types of products as a guide but ultimately you have to listen - for example how would you measure the size of a image or the extent of the soundstage?

That's why Brystons moto is "The Demo is Everything"

james

[GrooveControl]

If you have attempted to 'update' your system with more modern loudspeakers and found it wanting I would be interested in an opinion of what brought you to your current choice and what the mini dsp and eq give and take away from the enjoyment.

First off,  let me just say I'm not a golden eared audiophile.  Instead, an aging dude who loves the old stuff.  Parked my 3B SST for a bit, and giving my Marantz 2245 some exercise, and not really hearing much difference between them at the levels I listen at. The Bryston has like 20 times less distortion, but I don't hear it.

For many years, my main speakers were Energy 22 Pros that I've owned since the 80s.  My newest speakers are the Polk RTi A7 and KEF LS50. As you might expect, none of the Energy, Polk or KEF that I have can match the efficiency, dynamics and extension of JBL 4430. To my ears, the KEF LS50 is the biggest disappointment.  The bass is okay for such a small driver, but the mids are too forward, and the top end is too dull for me.  With help of EQ, graphic or DSP, I can make the LS50 sound great in the small TV room where I use them.   The Polk sound better than one would expect for their price.  Apply a bit of low bass lift and they are excellent.  Finally, the Energy, probably I best tweeter I've ever owned, great bass, but probably a bit too warm in the mid bass. In all I have 12 pairs of speakers including several DIY, and in all cases I prefer to use equalization. Partially for room node correction, and partially just for my listening preference for slightly elevated bass and treble.   

I could never understand why the EQ died. Most of today's amps don't even allow you to incorporate one in the signal path.  Disgusting imho.  Maybe because when people use an eq properly they jump off the upgrade wagon? Who knows.  Anyway, don't want to start a flame. I've always used EQs, and am moving into the DSP realm now.  Some people claim eq kills the soundstage, I never experienced that.  The best graphic eq I had was a KT DN360. I'm currently running a miniDSP 2x4HD in the tape loop of the Marantz 2245 and loving it.  With phono and tuner, the less than perfect A/D --> D/A conversion is not noticeable.  My digital source goes directly to the miniDSP and only goes through D/A conversion once.

Have fun.       

[R. Daneel]

I tend to think of EQ as a tool to improve upon an already reasonably well-tuned system. It will not correct for major flaws but it can alter the sound slightly to make it more in-line with the direction you want to take. Sort of like a mastering engineer who gets an award if the record sounds excellent and blames the recording and mixing engineer if it doesn't. That's EQ IMHO!

Over the years, there have been multiple threads on this topic but one stuck in my mind. One guy bought an expensive system and was very dissatisfied with the bloated bass he got. I don't know what he paid for the stuff but the preamp alone was over 10k. He complained to the dealer who came out with a brilliant idea that it must have been the preamp and the cables. He then sold him a different preamp and cables that cost even more. James Tanner was the first to reply to this thread saying that it's a classic case of a complete mismatch between the size of the room and the speakers. It's also a classic case of someone who has money but doesn't know how to spend it and a dealer who's mantra is 'to take your money before you hurt yourself with it'. No EQ will fix this!

But this pursuit of flat frequency response is tricky, both in attaining the desired goal of actually having a response that measures flat but that also sounds musically accurate. People rarely go to concerts these days and no, I'm not talking about spending a fortune on a yearly philharmonic subscription, but concerts with sometimes famous, sometimes less famous artists, held in smaller venues, city sports halls etc. It can be a rewarding experience to hear an unknown pianist playing a Yamaha C3 piano in a decently sized venue. You can hear the artefacts of the room but you can also hear the harmonic richness of the instrument, derived from both the harmony being played but also from the instruemnt's inherent character, one of the 'features' being the sympathetic string resonance. Is this 'character' reproduced correctly by the speakers? Do people even listen for such detail? I don't know. Talking about pianos, they haven't changed much for hundreds of years and they're still constructed to naturally supress the 5th and 7th harmonics because piano makers have discovered that these typically cause listening fatigue. Now imagine this - piano makers of those times certainly didn't know anything about acoustics, let alone the mathematical models on which modern acoustics is based on today, but they did learn to listen and to construct the instruemtns to make them pleasaing.

Being a structural engineer, I tend to think of it like this - I can design two completely different-looking road bridges that have the exact same span, resistance to load and serve the exact same purpose. Which one is the 'better' one? You cannot argue that science isn't at the heart of engineering and yet, there is the creative freedom in these mathematical equations so that there is more than solution, all of which are correct.

I should have been a preacher, I know But the point is, don't believe everything you see on YT or in the HIFI magazines. Go to as many live events as you can, go to the park and listen to the birds, listen to spoken word on the radio, and then decide how your speakers 'translate'. Educate yourself on how to listen and what to listen for and if you're musically inclined (or even if you're not), learn to play an instrument, even it it is an 8-tone Hohner flute or a mouth harmonica. At the very least, it will enrich your life and perhaps you'll even enjoy your music more and discover you already have everything you need, or ease your task of finding something that is indeed better because you can hear why, not because someone else said so.

Cheers - Antun

[GrooveControl]

Very well put Anton. 

EQ is indeed a tool, and in the world of hifi, a highly under-rated tool. 

The +/- 3db design target for speakers has been around forever and is very reasonable. It's where those pluses and minus happen in the range that allows manufactures to express their opinion of what good sound is. 

The 123s of buying your forever speakers are:
1 - will it play loud enough for you?
2 - will the bass go low enough for you?
3 - do you love how it looks? 

Those three things you cannot change. Leave the rest to the room you put them in, and an Equalizer.

EDIT: I should mention I'm talking about decent speakers here. No EQ will fix poorly designed speakers with resonance or XO design issues. 

[Blueshound]

Hi Why

I just ran now a FFT response of my set-up at the listening location for you.



james

James, I wonder how few people realize not only how near to perfect that curve is, from the perspective of music listening in the real world, but how VERY difficult to achieve with actual speakers in real, non-anechoic, rooms, without EQ above, say, 100Hz. Without EQ. Most speakers (i don't care what technology or cost) never come anywhere close in actual listening rooms. And I've heard many and sometimes even measured a few.

Having said the above, there are other equally important considerations in speaker setup relative to a given room. I find very small differences in toe-in, listener seating distance, space between the speakers relative to the principal listening location, distance of the speakers and listener from room boundaries, all to impact soundstaging, perceived sound response, perceived dynamic shadings, and overall musicality.

Cheers
Brian

[AllanS]

It took some time before I understood that the phrase “head in a vice” meant you couldn’t move your head without having some impact on what is heard.  I came across the phrase in the context of tweeter type.  But all the various discussions I’ve come across recently about adjusting speaker placement in 1/4” increments and careful toe in,etc, etc seems to suggest the sweet spot is really just a spot. Is perception really all that sensitive?

[whydontumarryit]

I now wonder what is the relationship between that text book in room response and the ability to move from the preferred listening location while maintaining a stable image, aka a large sweet spot. That is, does a flat response tend to stay that way at different listening positions and does a 3db variation turn into a 6db variation at some point vs listening position, making a flat response the primary goal.

The choices made (intentional or not) to reproduce an Olive-Weilti type, experienced listener preferred in room response (as indicated in the graph) would be one reason for visitors to be impressed.
Also, if the graph is showing an unequalized LF response, the boost at 25hz is a full octave below what is typically seen in your average listening room. What's going on? Is it the room dimensions responsible for the LF room modes or is it possibly parasitic vibrations from the 'live' walls in an average room. It would be interesting to see a comparison of high SPL LF thd measurements between each type of room. I suspect that the 3-10% thd usually seen below 100hz will mostly disappear in this correctly optimized room. I have never understood why the increased distortion is there in the first place and why nobody ever questions where it's actually coming from.

The original question about whether I should abandon this non-sensical purist attitude toward using EQ to get a flat response in my room was answered by the advantages claimed using active crossovers. If a linear phase, FIR, DSP device can be used for that purpose then it can be used to do what I want without concern. Used either way it's EQ, plain and simple.
Since the LW and DI of the middle-t are correctly optimized to take advantage of EQ without a problem, why not. There is also the uniform opinion that any changes made below the Schroeder frequency have no detrimental effects. So, it looks like a mini dsp is in my future.
All I need now is that BAX-1 config. file for the middle-t from Bryston to load into the mini dsp. Fat chance, eh. 

[jcn3]

James, I wonder how few people realize not only how near to perfect that curve is, from the perspective of music listening in the real world, but how VERY difficult to achieve with actual speakers in real, non-anechoic, rooms, without EQ above, say, 100Hz. Without EQ. Most speakers (i don't care what technology or cost) never come anywhere close in actual listening rooms. And I've heard many and sometimes even measured a few.

Having said the above, there are other equally important considerations in speaker setup relative to a given room. I find very small differences in toe-in, listener seating distance, space between the speakers relative to the principal listening location, distance of the speakers and listener from room boundaries, all to impact soundstaging, perceived sound response, perceived dynamic shadings, and overall musicality.

Cheers
Brian

I agree -- my mouth was hanging open looking at that!

[James Tanner]

I agree -- my mouth was hanging open looking at that!

Hi folks,

I guess my point is that with properly designed speakers and careful consideration of speaker placement and listening position you can achieve a lot without having to revert to systems (EX - EQ) that in fact negatively affects what has been the main design criteria of our Bryston speakers - namely as flat as possible sound power.

best
james

[GrooveControl]

Hi All,  I'm trying to understand why you consider a response that falls 10db from 1khz to 10khz, and then another 10db by 20khz to be near perfect?  Is this simply an alignment with your preference? A power response?  or what?  Not challenging anything here, just trying to understand it.  The chart is 10db per division. 

Thanks

EDIT: Okay,  I think Blueshound and jcn3 are referring to the first chart in this thread, and not the green one I'm including below.  Makes sense now.