Has Jim Smith discovered a magic formula for placing speakers correctly?

signal processing is bad for signal purity and gets further away from the ideal of "straight wire with gain".

One would like to think that with all the time & advances in EE, that the theoretical rule is not as hard & fast as it may have formerly been.
But, that said, DSP is an awful lot of signal juggling.
These considerations come up in 'tone control' discussions pretty often.
I guess it depends on the room, the equipment & the ears...... as it usually does.
 

It really depends what type and how much DSP is used, as with most things in almost any area of endeavor. A blanket statement either for or against DSP flies in the face of experimental and anecdotal data, as well as logic and common sense.
 
Would a hard-core hiend pure analog stereo hi-fi addict (TT & Tapes) dare to digitize his sound?
It's like taking pictures with a digital camera.

Enjoy Jim Smith and enjoy the music ...
Understanding Digital Room Correction For Audiophiles Article By Mitch Barnett Of Accurate Sound.

That article is an excellent yet relatively simple to understand description of how room correction DSP works and how complex the process is. I've not yet read Part 2 but I'm hoping it will describe what happen to high frequencies (those that room acoustics don't spoil) by this complex signal processing software.

It is my belief that, if the DSP is applied to full-range amps (as in most systems), then the higher frequencies suffer from the processor even though no effort is made to adjust the response of these higher frequencies. The result is a loss in lifelike reproduction and the excitement factor that we crave from music is diminished - at least if we use high quality speakers where we expect this excitement.

In many cases, particularly systems with subs or multi-speaker AV systems, the advantages of DRP may outweigh the disadvantages, but for simple 2-channel systems using high quality speakers with no subs, the loss of top end finesse may reverse the advantages of smoothing out the bass.

What do others think?
 
I think yours is one man’s opinion just like mine is.

I am a big believer in its benefits for my current room.

You appear not to be. Cool.
 
What if in some audiophile rooms only the frequencies below 250Hz are corrected while above that range they are untouched and simply fine-tuned by room treatments and furniture repositioning, including the gear, the couch, the paintings on the walls and the speakers.

Other rooms might require room correction where you can go higher than 250Hz. You adjust only the frequencies that are anomalies. The rest are left alone all by themselves in their purist form of high fidelity sound reproduction ... the music recordings from LPs, tapes, CDs, streams, AM/FM radio.
 
below 250hz, room correction can reduce the peak but cant fill the null. it will only stress your amp and creating noise.
 
What if in some audiophile rooms only the frequencies below 250Hz are corrected while above that range they are untouched

If you consider what happens with those higher frequencies, your will answer your own question. If you have your room correction DSP built into your full-range amp, then how can the upper frequencies avoid it? Only if you split out the upper and lower frequencies first and then send the bass to one amp with DSP and the rest to another DSP-free amp can the upper frequencies be preserved unaffected.

I agree there are instances where this processing may be of greater benefit than not, but I'd suggest that most 2-channel systems can be set up via other means - correct choice of speaker type for the room, then actual speaker model capable of delivering the bass you need, precise positioning, toe-in, perhaps tilt, furnishings, carpets, curtains, etc all should be exhaustively considered before resorting to DSP. It should be the last resort and not (as in many cases) the first attempt at getting best sound from a system. Jim has been silent on this thread and I don't blame him, but I would love to know how far up or down the list of things to do in ideal setting up that he would place DSP.
 
well, you dont treat your room and then add dsp on top.
dsp is for a problematic room when you DONT treat it otherwise
 
What is an audiophile?
_____

In my spare time I occasionally surf the internet ...

• Umlautica - Hear Hear!

"I've been using DSP for more than 10 years now and here's what I've learned.

TL;DR: placement > treatment > room correction

Conceptualize is the schroeder frequency of your room. The schroeder frequency is where modal effects determine the steady state response at the listening position. Think of it as the crossover point between (a) the frequencies where the room modal effects dictate response and (b) frequencies where the direct response of the loudspeaker. It's typically 200-400Hz. This image helps explain.

Consider DSP strategy as two logical categories:

Loudspeaker correction: the nearfield, quasi-acoustic, direct magnitude response of the speaker.

Room correction: primarily the frequencies below the Schroeder frequency of the room. It also will correct the room power as well.

The room power response is different from the magnitude response of the system. The room power is what happens when flat on-axis loudspeakers bounces bounce sound around and have it absorbed by the room. This is what Harman found in steady state measurements.

There's a saying that "you mostly hear the room". This usually omits that it mostly applies below the schroeder frequency. Above this frequency, the principal determiner of sound quality is the direct loudspeaker response. Below it, the room takes over.

DSP is a great tool to have, it's not a panacea. Tools can be misused.

For example, if your speakers have poor directivity (off-axis response), DSP cannot fix it. You can dump gain on a null and do nothing but add distortion and eat headroom. DSP cannot fix the null but it can help by taking some energy out of a peak from a room mode.

Some generalizations of what to expect for correcting above the schroeder frequency:

Bad on-axis: easy DSP gains - fire and forget

Good on-axis with decent directivity: non-trivial DSP gains - use caution

Great on-axis with good directives: difficult to improve with DSP - can't fix what isn't broken

So if someone has JBL M2 towers, it's usually unreasonable to assume that full-range correction will improve things. It can do more harm than good. The M2 ship with an amplifier preloaded with DSP filters that provide a flat on-axis response.

My ideal order of operations:
Place subwoofer/speakers where room mode issues are minimized at the main listening position. For subwoofers, try the "sub crawl". Try midpoints of the front and rear wall. Try nearfield subwoofer placement. Try corners. Try everything until you're ready to find a new hobby. If you have a rectangular room, the REW Room Sim can offer tips.

Set up your room treatment. To make a long topic short; first reflection points within ~5ms of path length, wideband treatment behind to reduce SBIR, and dispersion on the back wall.

Use DSP to time align each subwoofer with your main speakers. The REW Spectrogram is a fantastic tool for this task.

Set subwoofer gain to be a few dB above your main speakers.

Experiment with the main/sub crossover point. If you have a room mode below 80Hz, it may help to center the room mode at the crossover point.

Find what you prefer for toe-in of your main speakers. This depends on your speakers and boundary reflections. Get a friend to help. If you don't have any friends, try moving your head forward or back at your listening position to inform your next adjustment.

Measure your loudspeakers using gated near-field measurements to find the direct response at the acoustic center of the loudspeaker.

Use low-q filters to correct your loudspeakers to a flat response. Perfection is not the goal. Don't fuss with narrow dips. Save corrections below ~250Hz for room correction.

Now you're ready for room correction!

Run your favorite tool; DIRAC, Audessey, REW, GLM, etc. Be sure to take multiple measurements from multiple points to get an averaged spatial response. Correcting for one point in space almost guarantees that you'll make corrections that are worse for the space around that point.

Be careful trying to extend your bass response below the natural response of your subwoofer. It can take more excursion or power than the subwoofer can provide.

You may find that your room correction software performs best when low passed a little above the schroeder frequency of your room. Calibration can be limited to frequencies such as 20Hz-1000Hz like this. You also may find that a good target response follows the measured room response of the top octaves of your speakers.

You may find that room correction sounds thin. Many prefer a 3-6dB bump around 50-60Hz helps fix this.

Have fun with it. Try things. Have reasonable expectations."
_____

Is an audiophile a person ready to explore the multitude ways to improve his sound in his room? ...Like a scientist in search; by testing and measuring and analyzing data and ear listening and correlating and understanding and amelioring and gathering graphs and writings and observing and more testing and more listening and experimenting and discovering along the way.

Loudspeaker's positioning in a room equipped with a hi-fi stereo sound system is one of them ... Jim Smith. No matter the music source (albums recorded on LPs, CDs, Audio Streams, R2R Tapes, V8, ...) @ 24, 48, 96, 192, 384, 768Hz.
 
well, you dont treat your room and then add dsp on top.
dsp is for a problematic room when you DONT treat it otherwise

DSP is no substitute for proper speaker setting up. There's no point in buying costly speakers, then not bothering with getting them set up correctly and similarly there's no point in putting good speakers into an echo chamber of a room or any roon where no thought is given to carpeting and curtains, etc. Getting these right is far more rewarding than chucking DSP at the problem, hoping they'll go away!
 
OK, here goes. Jim Smith's "magic formula" for placement of speakers in relation to the listening chair is stated as 83%.

Now, as I understand that, if the distance from each tweeter to the listener's ears (Y in his video description) is 100 units, then the tweeters should be 83 apart (X in his description). X divided by Y is the magic 83%

I said I'd work out the angle this represented in degrees - ie the angle between lines drawn from the listener to the 2 speakers.

My calculation is that this angle is ideally 49 degrees.

If you need the geometry, here it is. By using an isosceles triangle with the base being the distance between the speakers at 83 units and the equal sides as 100 units, the angle between your ears and the 2 speakers is 49 degrees.

Use a compass or protractor to measure your angle to see how well your speakers are placed compared with Jim’s formula.

My own speakers are too close together, but unfortunately it’s not realistic to move my speakers about 2 feet further apart. Equally I’d have problems moving the speakers, or a huge L-shaped sofa (that’s over 20 ft long) forward 3 feet. I’ll have to be content with 68% or 40 degrees from my listening seat. Peter
 

Attachments

  • Jim Smith's Magic Angle in Degrees.PNG
    Jim Smith's Magic Angle in Degrees.PNG
    34.8 KB · Views: 85
Where did the .83 number come from? I know it’s what Jim Smith recommends, but was it derived from some sort of calculation, or is just his preference? Apologies for being too lazy watch videos. FWIW, I’m at about .9, for no particular reason other than it works me.
 
Where did the .83 number come from? I know it’s what Jim Smith recommends, but was it derived from some sort of calculation, or is just his preference? Apologies for being too lazy watch videos. FWIW, I’m at about .9, for no particular reason other than it works me.
It comes from Jim setting up dozens, perhaps hundreds of rooms.
 
I don't see the point of calculating the angle, working with distance so much easier and accurate
 
FWIW...

I use the 83% as a starting point so that I don't lose time, especially after having voiced over a thousand systems.

I was pleased when Wilson came out with their 1/1.2 set-up ratio years ago. 1 divided by 1.2 = 83.33%!!!

If a set-up is having issues in a certain room, from time to time I will check the ratios to help me understand what might be happening.

If others want to use other set-up info, more power to them! I hope it all works out for them.

As recently as yesterday, I had a very knowledgeable and perceptive RoomPlay Reference client here for a listening/learning session, and like most visitors, it was the best system set-up he had heard. I'm talking about Musical Engagement, not audiophile sound effects...

This with a system that costs less than the power cords from the last system...

With all of these comments floating around, I did check the set-up - it was between 83.4 & 83.5%.
 
I think if audiophiles did three things as a start to their system setup, they would be in very good shape.

1. Pull speakers 5’ or more from the front wall
2. Choose the listening position that has the smoothest frequency response under 250 hz
3. Start with Jim’s 83% ratio for speaker placement and tweak from there by ear

I don’t have nearly the experience that Jim does, but enough to know that this can lead to very engaging sound. I thank Jim for providing the guidance to help us all get there.
 
I think if audiophiles did three things as a start to their system setup, they would be in very good shape.

1. Pull speakers 5’ or more from the front wall
2. Choose the listening position that has the smoothest frequency response under 250 hz
3. Start with Jim’s 83% ratio for speaker placement and tweak from there by ear

I don’t have nearly the experience that Jim does, but enough to know that this can lead to very engaging sound. I thank Jim for providing the guidance to help us all get there.

#1 is questionable; either Wilson's or Jim's method should probably be used instead (FWIW my speakers are 58" from the front wall; 64" - where I started - is/was definitely too far out into my 23' room)
 
#1 is questionable; either Wilson's or Jim's method should probably be used instead (FWIW my speakers are 58" from the front wall; 64" - where I started - is/was definitely too far out into my 23' room)

Jim definitely recommends pulling the speakers out into the room. I used the 5’ minimum suggestion as a basic starting point.
 
Back
Top