Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: How do you know if your sound is right? (Read 19008 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

How do you know if your sound is right?

Reply #50 –
But I want to know when flat is desirable given all the criteria you have listed and when it is not desirable given all the criteria listed.

How do you know if your sound is right?

Reply #51 –
Quote from: Rich B on
But I want to know when flat is desirable given all the criteria you have listed and when it is not desirable given all the criteria listed.



Rule of thumb: Flat in the near, non reverberant field;  non-flat (rolled off high end) in the far, reverberant field.

How do you know if your sound is right?

Reply #52 –
Quote from: Arnold B. Krueger on
Quote from: Rich B on
But I want to know when flat is desirable given all the criteria you have listed and when it is not desirable given all the criteria listed.



Rule of thumb: Flat in the near, non reverberant field;  non-flat (rolled off high end) in the far, reverberant field.


Okay.

Are there any studies to corroborate this? I know Sean Olive has a study about user preference and he says that people prefer smooth response as opposed to hocus pocus response (topsy turvy) which I found interesting. I was expecting the Martin Logins to score highly but they did the worst in the blind testing. The response looks very sub-optimal for that speaker.

How do you know if your sound is right?

Reply #53 –
Arny, can you explain the advantage of an anechoic-flat speaker playing in a non-anechoic space as opposed to a non-flat speaker measured in an anechoic space playing in a non-anechoic space?

I can only think of one thing - variance in frequency responses. Are there any other advantages?

How do you know if your sound is right?

Reply #54 –
Quote from: Rich B on
Arny, can you explain the advantage of an anechoic-flat speaker playing in a non-anechoic space as opposed to a non-flat speaker measured in an anechoic space playing in a non-anechoic space?

I can only think of one thing - variance in frequency responses. Are there any other advantages?



The problem I have with the question is that virtually every real world speaker is decidely non-flat when measured in a non-anechoic space unless the measurement technique is trying to isolate anechoic response.

Another problem is that the alternatives are too broadly stated.

So the advantage that was questioned pretty much never actually exists in a useful way.

For example take the Primus P363 which Harman has shown with blind tests and measurements to be some kind of a giant killer and highly preferred in listening comparisons to far more expensive and often highly regarded alternatives.

The P363 is pretty close to flat on-axis. Doing that is relatively cheap and easy.  Its off axis (reverberant field) response is downward tilted. I think both response curves were completely intentional. 

Any serious compromises to keep costs down were made in the reverberant field response, but they were minimized and are in fact less than the compromises found in any of the far more expensive speakers that it was compared to.

The weakest part of the P363 is IME  the tweeter. I had 2 of 5 samples fail in use within a year. The non-exact replacements have a little flatter response and a little more treble extension, but in our testing we found that they could take a lot more power and play louder without sounding harsh.  Technical tests show that the P363 and its Primus relatives have very competitive dynamic range in the frequency range of the rest of the drivers.

My experience with the P363 is that I upgraded them to PC351s which had narrower response patterns in the frequency ranges below the tweeters.  I then upgraded them to LSR 308s which had even narrower dispersion in both the midrange and the treble.  My listening room is pretty reverberant. and listening distances aren't short.  The LSR 308s seem to have more dynamic range at all frequencies and have held up just fine thus far. My midline AVR now drives relatively high value resistors (about 1,000 ohms)  instead of low impedance (< 4 ohms) speakers.  Doesn't necessary help, but sure doesn't hurt.

How do you know if your sound is right?

Reply #55 –
Quote from: Rich B on
Arny, can you explain the advantage of an anechoic-flat speaker playing in a non-anechoic space as opposed to a non-flat speaker measured in an anechoic space playing in a non-anechoic space?

I can only think of one thing - variance in frequency responses. Are there any other advantages?

One will sound a lot more 'right'. 

But it is more complicated than that. Look at the ~8 month old thread I linked earlier. Search for "direct sound", "sound power" ...
"I hear it when I see it."

How do you know if your sound is right?

Reply #56 –
Quote
I know Sean Olive has a study about user preference and he says that people prefer smooth response as opposed to hocus pocus response (topsy turvy) which I found interesting.
I find that reassuring.

How do you know if your sound is right?

Reply #57 –
interesting != understanding, let alone accepting.

That we're having another trip on this merry go round should be enough evidence of that.

How do you know if your sound is right?

Reply #58 –
From my perspective, if we forget about speakers then for an actual instrument or voice in a room the direct sound will be "flat" and the the reflected sound will vary with room response and the off axis response of the sound source.

How do you know if your sound is right?

Reply #59 –
Quote from: drewfx on
From my perspective, if we forget about speakers then for an actual instrument or voice in a room the direct sound will be "flat" and the the reflected sound will vary with room response and the off axis response of the sound source.


Exactly. But we've already tried explaining the concept of direct sound etc. to Rich B 8 months ago. Not sure he ever understood, but I'm quite sure that he never read the resources he was pointed to.
"I hear it when I see it."

How do you know if your sound is right?

Reply #60 –
Quote from: drewfx on
From my perspective, if we forget about speakers then for an actual instrument or voice in a room the direct sound will be "flat" and the the reflected sound will vary with room response and the off axis response of the sound source.



The reflected sound is pretty well guaranteed to have a rolled off high end because most things that reflect are increasingly lossy at higher frequencies due to both absorbtion and diffusion.  The reflected sound is not only rolled off, but has a lot of bumps due to standing waves, resonances, etc.

How do you know if your sound is right?

Reply #61 –
Quote from: Arnold B. Krueger on
Quote from: dhromed on
Quote from: mzil on
How do you know if your sound is right? Easy, when it is completely indistinguishable from the original source.
  What original source? There is no original source for a band with electric instruments in a studio, or electronic music, since we can only access that sound filtered though speakers and other hardware— which is the very thing we're discussing.
  The original source for an electronic instrument is usually defined by the musician playing the instrument - it is the sound of him playing his preferred electronic instrument through his preferred speakers in his preferred performance space.  The original source for a musical work that was created in the studio is what the mixdown technician heard in his monitoring system as he mixed it down into what he thought was the final mix. This may differ from what the mastering engineer created.
  I've always considered it absurd to use electric instruments to evaluate the tonal accuracy of things like headphones and speakers because unlike true acoustical instruments, say for example flutes, which vary only slightly from brand/model to brand/model [and I wouldn't be surprised if the top ones actually are nearly indistinguishable in blind studies]  electronic instruments such as the electric guitar can vary drastically, even when using the very same model, due to things like the musician's whimsical setting of the instrument's front panel tone controls. There is no absolute sound to an electric guitar; it is instead arbitrary, whereas flutes always sound like flutes.



Yes, there will be subtle differences in the flute recording's mics, micing, performance venue room reverberation characteristics [if not recorded anechoically], etc., not to mention those of the playback room, however a flute does have an intrinsic sound, unlike an electric guitar which varies all over the map by many dB.

Using electronic music to evaluate speaker tonal accuracy is like using CGI generated cartoon images of, for example, Shrek to evaluate color accuracy on a video monitor. Shrek's skin color was just the whimsical setting of some CGI artist we can only guess at. This is why video calibration, when done by eye, uses images of real world items, often fruits, vegetables, and people, which although they vary slightly have a basis in reality and aren't just some arbitrary setting on a control knob.

How do you know if your sound is right?

Reply #62 –
Quote from: mzil on
musician's whimsical setting of the instrument's front panel tone controls.

You must not be a guitar player.

"tone is in the fingers"

How do you know if your sound is right?

Reply #63 –
Quote from: Rich B on
Are there any universally accepted studies that suggest a flat on-axis response in a room is desirable for good sound quality? Or a flat on-axis response is desirable assuming the room is reverberant? Or dead? Or ..?


There are no universally accepted studies about anything.  You are asking for something neither science nor engineering can ever deliver.
Ed Seedhouse
VA7SDH

How do you know if your sound is right?

Reply #64 –
Quote from: mzil on
electronic instruments such as the electric guitar can vary drastically, even when using the very same model, due to things like the musician's whimsical setting of the instrument's front panel tone controls. There is no absolute sound to an electric guitar; it is instead arbitrary, whereas flutes always sound like flutes.


The same is true of any pipe organ going back a thousand years or more...
Ed Seedhouse
VA7SDH

How do you know if your sound is right?

Reply #65 –
Quote from: Ed Seedhouse on
Quote from: Rich B on
Are there any universally accepted studies that suggest a flat on-axis response in a room is desirable for good sound quality? Or a flat on-axis response is desirable assuming the room is reverberant? Or dead? Or ..?


There are no universally accepted studies about anything.  You are asking for something neither science nor engineering can ever deliver.


On balance most published reviews of new and highly regarded speakers show on-axis response that is pretty close to being as flat as possible.  The off axis response generally has a smooth rolloff that increases with angle.

I believe that this is the latest speaker review on the Sterephile web site, and like so many before it we see a speaker with approximately smooth and flat on-axis response, with response falling off at an increasing rate as the measurements move off-axis:

http://www.stereophile.com/content/tekton-...er-measurements

How do you know if your sound is right?

Reply #66 –
The big thing with all these correction systems that are out there is that you DO NOT correct to flat at listening position , you need a bass hump and a treble droop which makes it mandatory to use a house curve - ie a curve that is not flat.

How do you know if your sound is right?

Reply #67 –
You still seem to be confusing on-axis with sound power etc.
"I hear it when I see it."

How do you know if your sound is right?

Reply #69 –
Quote from: Rich B on
Quote from: xnor on
You still seem to be confusing on-axis with sound power etc.


How so?



In listening rooms, the usual convention is to try to get on-axis response pretty much flat, and that will naturally give you the off-axis droop.

Of course your preference may be not for flat response, but my system is calibrated for flat on-axis response and it seems to work out pretty well.

How do you know if your sound is right?

Reply #70 –
Quote from: Arnold B. Krueger on
Quote from: Rich B on
Quote from: xnor on
You still seem to be confusing on-axis with sound power etc.


How so?



In listening rooms, the usual convention is to try to get on-axis response pretty much flat, and that will naturally give you the off-axis droop.

Of course your preference may be not for flat response, but my system is calibrated for flat on-axis response and it seems to work out pretty well.


As opposed to? A flat sound power response at the seat?

How do you know if your sound is right?

Reply #71 –
Rich, consider the following frequency response charts:

“The Subjective and Objective Evaluation of Room Correction Products” Audio Musings by Sean Olive: http://seanolive.blogspot.ca/2009/11/subje...luation-of.html

In the Olive paper, the preferred response is smooth from 20 Hz to -10 dB at 20 kHz. See slide 24:




In 1974 B&K wrote an interesting paper http://www.bksv.com/doc/17-197.pdf called, “Relevant loudspeaker tests in studios in Hi-Fi dealers' demo rooms in the home etc. using 1/3 octave, pink-weighted, random noise” From their research, the preferred in-room frequency response at the listening position is flat to 200 Hz, -3db at 2 kHz and -6db at 20 kHz.See Figure 5 in the paper:



In 1998, the European Broadcast Union produced a Tech note (EBU-Tech 3276) called, “Listening conditions for the assessment of sound programme material: monophonic and two–channel stereophonic”. https://tech.ebu.ch/docs/tech/tech3276.pdf  See Fig 2 on Page 6:



If you search around, you will find similar guides/preferences from Dolby and BBC, plus anecdotal preferences from many folks who have played with equalizers and so called Digital Room Correction systems like Acourate and Audiolense.

In my case, here are 6 measurements taken around a 6' x 2' grid where my couch would normally be at the listening position:



I attribute the smooth frequency response across the listening area above 500 Hz to using constant directivity waveguides. 

That's my preference.  What's yours?

How do you know if your sound is right?

Reply #72 –
Quote from: Rich B on
I think the short answer is that you never do.

There is no reference you can tap , as no recording will ever sound like live and how close it comes is dependant on production , what any recording engineer hears is mangled by the control room and the monitors so might sound radically different in your room.

Even the sound you hear at listening position cannot be reconciled with it measuring "flat" as the room gain and treble droop associated with the room as well as any reflections mandate that the FR at listening position is most certainly and should NOT be flat , it is a taste based target curve.

Not even headphones can help , they may be accurate tonally , but there are other shortcomings they cannot overcome that only speakers deliver.
So it essentially all boils down to your own taste and your own ability to suspend disbelief that you are listening to a recording.

Thoughts?


I disagree . The point of engineering,  and I can tell you I most definitely want my monitors FR flat, is that your recordings sound the same, to you, on other  systems  on as wide s range of audio systems as possible.  Your mix is created to your taste but you want your taste to translate.

How do you know if your sound is right?

Reply #73 –
Quote from: Rich B on
Quote from: Arnold B. Krueger on
Quote from: Rich B on
Quote from: xnor on
You still seem to be confusing on-axis with sound power etc.


How so?



In listening rooms, the usual convention is to try to get on-axis response pretty much flat, and that will naturally give you the off-axis droop.

Of course your preference may be not for flat response, but my system is calibrated for flat on-axis response and it seems to work out pretty well.


As opposed to? A flat sound power response at the seat?


The listener hears the summation of the direct sound straight from the speaker, and the indirect sound that bounces around the room on its way to the ear. The direct sound first, the indirect sound delayed by the reverberation time of the room. The usual goal is flat response for the direct sound, and downwardly sloped indirect sound. This then corresponds to how we hear natural sound in a performance room or concert hall.

How do you know if your sound is right?

Reply #74 –
Quote from: Arnold B. Krueger on
Quote from: Rich B on
Quote from: Arnold B. Krueger on
Quote from: Rich B on
Quote from: xnor on
You still seem to be confusing on-axis with sound power etc.


How so?



In listening rooms, the usual convention is to try to get on-axis response pretty much flat, and that will naturally give you the off-axis droop.

Of course your preference may be not for flat response, but my system is calibrated for flat on-axis response and it seems to work out pretty well.


As opposed to? A flat sound power response at the seat?


The listener hears the summation of the direct sound straight from the speaker, and the indirect sound that bounces around the room on its way to the ear. The direct sound first, the indirect sound delayed by the reverberation time of the room. The usual goal is flat response for the direct sound, and downwardly sloped indirect sound. This then corresponds to how we hear natural sound in a performance room or concert hall.


And this is what Sean Olive and similar sources advocate?