What to bring to speaker listening tests?

[dbxkompress]

First of all, thank you for your attention.

Long time unregistered lurker, now fully joined because I'll be auditioning some speakers for new home.

So... how can we get *somewhat* objective measurements for these tests?

I'm thinking a checklist like this, please add more suggestions:

1. Laser measurement for distance control
2. Some kind of mic for SPL measurement and frequency response. Obviously this is affected by room acoustics but I could at least see "oh this is what I'm hearing, not just my imagination.

Things I'd like to measure but not sure how:

• Distortion?
• H and V dispersion?
• Maybe some ways to test the room acoustics? Like anyway to "tell" what the room sounds like? Use test tone?

Anything else you'd suggest?

Most likely I'll be bringing my laptop to a private listening test, so anything I install on the laptop would be good. Obviously I'm not certain I could run setup measurements like using sweep tones or calibrate, but at least I could ask them to play the same tracks. Yes it's super not controlled, but the point is to quantify what I'm hearing, or try to catch what I missed (too much distortion).

Currently thinking of auditioning KH420, Kii Three, Mofi V10, Genelec 8361, if that matters.

 

[Keith_W]

Welcome to ASR.

There is no point taking measurements of the loudspeakers you listed since proper measurements have already been taken: KH120, MoFi, 8361. Any in-room measurements you take in a showroom will be very different to what you will get when you buy one and take it home, anyway - including any measurement of the room acoustics.

I would simply go and listen at a normal listening volume. Bear in mind that if YOUR room is bigger than the demo room, or if you plan to listen from further away, you will need a speaker that can go louder without sounding strained.

 

[dbxkompress]

Thanks @Keith_W
One reason I'd like to see some measurements is also to compare expected vs realistic - some gear might be purchased used.
Also there's no measurement on Mofi V10 yet... and also that list of speakers I mentioned aren't exhaustive. For example, some random used Focal Utopia models, or the newer Perlisten R/S series. I saw an used S9900 and wanted to see if it was maintained well too.

And, sorry for repeating, it's not to "measure for objectivity", but "measure for reference" - "this sounds hollow to me, is the mid really depressed or just imagining it...?"

But of course, if there are other measurements other than frequency, i.e. distortion, cabinet resonance, etc, that'd be nice too.

In other words, I'm picking THAT PARTICULAR piece of gear, not just "in general this is what I expected, give me one of them and hope the quality assurance does its job"... If that makes sense.

 

[Duke]

In addition to @Keith_W's suggestions, I suggest one additional listening test:

Turn the volume up a bit louder than normal, walk out of the room, and listen through the open doorway with no line-of-sight to the speakers. All you will hear from outside the room is reflected sound. If it still sounds good from outside the room, this indicates the reflection field has a good spectral balance, and ime this bodes well for long-term fatigue-free listening. (Ime a significant discrepancy between the spectral balance of the direct sound and the reflections correlates with listening fatigue setting in after a while.)

This is sometimes called the "Listening In Another Room", or LIAR, test. If the room the speakers are in has a lot of absorption, this test is not reliable because the high frequency energy will have been largely removed from the reflections.

 

[dbxkompress]

In addition to @Keith_W's suggestions, I suggest one additional listening test:

Turn the volume up a bit louder than normal, walk out of the room, and listen through the open doorway with no line-of-sight to the speakers. All you will hear from outside the room is reflected sound. If it still sounds good from outside the room, this indicates the reflection field has a good spectral balance, and ime this bodes well for long-term fatigue-free listening. (Ime a significant discrepancy between the spectral balance of the direct sound and the reflections correlates with listening fatigue setting in after a while.)

This is sometimes called the "Listening In Another Room", or LIAR, test. If the room the speakers are in has a lot of absorption, this test is not reliable because the high frequency energy will have been largely removed from the reflections.

Woah, many thanks @Duke !

This seems like a good test surprisingly for non-hyper treated rooms, which is quite many of them. I tried walking around the room before and at different heights, but never in a different room. Fatigue-free... maybe some combined factors.
I wonder if that's why omni-directional speakers are many people's favorite but measures poorly?

 

[DanielT]

First of all, thank you for your attention.

Long time unregistered lurker, now fully joined because I'll be auditioning some speakers for new home.

So... how can we get *somewhat* objective measurements for these tests?

I'm thinking a checklist like this, please add more suggestions:

1. Laser measurement for distance control
2. Some kind of mic for SPL measurement and frequency response. Obviously this is affected by room acoustics but I could at least see "oh this is what I'm hearing, not just my imagination.

Things I'd like to measure but not sure how:

• Distortion?
• H and V dispersion?
• Maybe some ways to test the room acoustics? Like anyway to "tell" what the room sounds like? Use test tone?

Anything else you'd suggest?

Most likely I'll be bringing my laptop to a private listening test, so anything I install on the laptop would be good. Obviously I'm not certain I could run setup measurements like using sweep tones or calibrate, but at least I could ask them to play the same tracks. Yes it's super not controlled, but the point is to quantify what I'm hearing, or try to catch what I missed (too much distortion).

Currently thinking of auditioning KH420, Kii Three, Mofi V10, Genelec 8361, if that matters.

Hello newly registered member.

Is it in your auditioning room that you will then place the speaker you like best? That is, do you borrow the speakers and listen to them at your home?
If so, will the room be furnished in about the same way auditioning vs permanent?

Extreme case: Hardly any furniture at all, an almost completely bare room (lets say auditioning room), then you like speaker X best, but in the same room over furnished with lots of carpets and stuff here and there, then you like speaker Y best.

 

[GS430man]

Bring some copies of your favorite music - stuff you know really well.

 

[dbxkompress]

@DanielT Sadly no... I'lll be going to different places trying to guestimate how they'd sound in my listening room.
But I'll be working with my audio engineering friends to help treat my room as much as possible, so that the room would have less impact on the speakers.

And because very few people/comapnies carries all these model at once, I most likely won't be able to do ABX tests, or have any control over the form/room. Which is why I'm trying to do as much measurements I could so that I might refer to my notes and data later, instead of just notes.

That said, while talking to you I had an idea: maybe I can bring the same mic, very close to the speakers, to record at different positions of components (tweeter, woofer), and then use a test tone on the room with the same mic, so maybe that can give me some reference?

 

[DanielT]

Welcome to ASR.

There is no point taking measurements of the loudspeakers you listed since proper measurements have already been taken: KH120, MoFi, 8361. Any in-room measurements you take in a showroom will be very different to what you will get when you buy one and take it home, anyway - including any measurement of the room acoustics.

I would simply go and listen at a normal listening volume. Bear in mind that if YOUR room is bigger than the demo room, or if you plan to listen from further away, you will need a speaker that can go louder without sounding strained.

Tricky with demo rooms. I don't really know what conclusions you could draw if one is very different, in every way, compared to the room at home where you will then put the speakers?

But okay, you get to see the speakers physically and that in itself makes sense to do before buying.

 

[dbxkompress]

Tricky with demo rooms. I don't really know what conclusions you could draw if one is very different, in every way, compared to the room at home where you will then put the speakers?

But okay, you get to see the speakers physically and that in itself makes sense to do before buying.

Like I should be able to do some test with some of them in the same room (likely Neumann and Genelec because of the audio engineering friends), but others would not be controlled. Dirty testing... this is why I'm wondering what I could do given what little I could control, to avoid being just another "listen with your ears man, just trust the experience dude" thing.
Look, if I trust myself I wouldn't need measurements, okay?

 

[DVDdoug]

Bring some copies of your favorite music - stuff you know really well.

That's probably the best advice yet!

In the old days many shops would allow you to borrow the speakers (maybe more than one pair) to audition at home. I don't know if anybody still does that.

An SPL meter might be helpful so you can approximately level-match. Frequency response testing will probably take too long.

An SPL meter isn't for frequency response. Most SPL measurements are A-weighted to approximate the human ear and even "flat" is not necessarily flat enough for measurements.

Generally for frequency response you need a calibrated measurement mic and REW, but in-room measurements are most-useful for lower frequencies where the room plays a big part.

The REW website is full of information, but you are measuring the speakers in that room.

I doubt this will be possible or practical, and possibly "not allowed."

Distortion?

REW can measure distortion (if you have time for that stuff).

But I'm not sure it's that important to measure... If you crank-up the volume and hear distortion that's bad. If not, that's good!

Frequency response isn't so easy to do by-ear, especially if you can't immediately A/B.

H and V dispersion?

That's not really possible in the room.

Maybe some ways to test the room acoustics? Like anyway to "tell" what the room sounds like? Use test tone?

You can't separate the room.

Tones won't be that helpful for listening but you'll probably hear differences if you A/B pink noise (or white noise). But if you're not in the same room at the same time that's probably not useful either.

Amir has a video about how he measures speakers with his specialized-expensive equipment. IT' not something we can do "at home"

 

[DanielT]

@DanielT
But I'll be working with my audio engineering friends to help treat my room as much as possible, so that the room would have less impact on the speakers.

Not as much acoustic treatment as possible. It will be a boring listening room. As much treatment as suits you. Not too much. Not too little.

None of these rooms appeal to me for music listening, too little, too much:

@DanielT
That said, while talking to you I had an idea: maybe I can bring the same mic, very close to the speakers, to record at different positions of components (tweeter, woofer), and then use a test tone on the room with the same mic, so maybe that can give me some reference?

Why? They are already tested, see @Keith_W #2
In addition to that, what you say about recording the drivers with the same mic then a test tone with the same mic for reference, I don't really understand what you mean. Reference for what?

 

[kemmler3D]

One reason I'd like to see some measurements is also to compare expected vs realistic - some gear might be purchased used.

For this, you can run a sweep and listen for any obvious defects. Deviation from published measurements will tend to be extreme (i.e. you can hear it pretty easily), or negligible.

I think tones can be useful for on-the-spot testing to hear major problems, but with the speakers you're proposing to listen to, they will really only be good for identifying damage/defects in the hardware. Bring a recording of a sweep on your phone, or load up this site: https://www.szynalski.com/tone-generator/

this is why I'm wondering what I could do given what little I could control, to avoid being just another "listen with your ears man, just trust the experience dude" thing.

It's a good thought but this is 1) mostly redundant since we have good measurements of these speakers already (yes, including the sourcepoint 10) and 2) very difficult and awkward to do on the spot. I used to go around and demo speakers at factories and trade shows... I was never really in a position to do this kind of testing in the moment, and it wouldn't have helped that much if I did.

Let me put it this way... if you haven't measured a ton of speakers before, then the likelihood of you finding meaningful and genuine differences between these SOTA speakers on the spot using a mic is very low. They're already within a few percent of each other, and any small mistake in making the recordings will tend to create a larger deviation than there actually is.

My advice is this:

Listen to a sweep and pink noise on each speaker. Make note of anything that sounds off or broken.

Then listen to 10+ recordings you know really well on each speaker*. Make note of anything you notice, good, bad, or neutral. Try to listen to each frequency range one at a time - low bass, mid-bass, low-mids, mids, etc. Listen at different places in the room - standing, sitting, near, far, and LIAR as @Duke suggests.

Also make note of the acoustics in the room. You can get a quick sense by clapping your hands and listening to the echo.

Take your notes, go home, and look at the published measurements of the speakers. See where your impressions line up with the measurements. When you figure out where your likes and dislikes line up with the actual objective performance of the speaker, it should help you reach a decision.

*My suggestion is to use songs that work as "stress tests" if you have some familiar tracks that can do that. For example, speakers with too much distortion or exaggerated treble make Paul Simon sound sibilant on Diamonds on the Soles of Her Shoes, because his "S" sounds are already a bit strong. Speakers with bass resonances or lots of group delay sound really muddy on Gimme the Loot by Notorious BIG because there's a lot going on in the bottom 2 octaves. Etc.

 

[ahofer]

In addition to @Keith_W's suggestions, I suggest one additional listening test:

Turn the volume up a bit louder than normal, walk out of the room, and listen through the open doorway with no line-of-sight to the speakers. All you will hear from outside the room is reflected sound. If it still sounds good from outside the room, this indicates the reflection field has a good spectral balance, and ime this bodes well for long-term fatigue-free listening. (Ime a significant discrepancy between the spectral balance of the direct sound and the reflections correlates with listening fatigue setting in after a while.)

This is sometimes called the "Listening In Another Room", or LIAR, test. If the room the speakers are in has a lot of absorption, this test is not reliable because the high frequency energy will have been largely removed from the reflections.

So that’s what all the wives who run in from the kitchen are doing!

My own advice: Listen to each speaker for a while. If a speaker is going to be fatiguing, it should show up in 10-15 minutes.

 

[Nopiano]

Just take your credit card and get the Kii Three!

In my opinion no point in listening to any passive alternative. You might prefer other actives. Can you run to the Kii BXT as well? The only downside is the immense range of adjustments the Kii controller can make, best experienced at home. (Yes, I spent a day with them thanks to a wealthy friend)

If you insist on hearing the others then I’ve nothing much to add to what’s been said, except take some music you really don’t like much. Use it as a palate cleanser and you might be surprised at what it reveals. Good luck!

 

[goofytwoshoes]

OP, if you have not already read Floyd Toole's posts on this forum, and/or at avsforum.com, start reading. Also, posts by Sean Olive. You'll learn a lot more than you would by doing measurements in various "show rooms".

I can comment only on Genelec SAM speakers, which I have started learning-about and using several months ago. There is much good reading on the Genelec website, and many excellent videos that they have posted on YouTube. Have a look at the Genelec GLM Manual - it is over 100 pages, and most of it will not apply when you don't have their speakers and GLM Kit in front of you, but it will give you an idea of what goes in to their software system. Also, read the sample GRADE report on their website. GRADE reports for my speakers (8330A monitors, with one or two 7350A subs) are giving me an excellent overview of system performance and how to improve it.

 

[geickmei]

Damn, it pains me to have to keep saying this over and over again. There are things about speakers that you can measure and things that cannot be measured, only heard. You can measure all about axial response, power response, and even direct to reflected ratio (but you probably wouldn't think to do that one). But you can't measure spatial response - imaging.

Here is basically what to listen for: Listening in stereo, with correctly positioned speakers - pulled out from all walls - in a normally furnished room, not a highly treated anechoic chamber - the image plane should be heard somewhat behind the plane of the speakers and wall to wall, not just speaker to speaker. This effect of the speakers disappearing and all images coming from points in space rather than a clothesline stretched between the speakers is caused by a sufficient output to the rear and rear/sides which is in turn caused by the radiation pattern of the speakers. The simplest example I can give you with commercial speakers would be the open baffle, which has at least as much output to the rear as to the front - like the True Ribbon Maggies or the Quad ESL-63 if they haven't put some dampening curtain on the back. The open baffle should have dipolar tweeters with equal output front and back, not some directional horn tweeters. All of this is caused by a decent ratio of direct and reflected sounds into the room, which audible effect cannot be measured directly and which I call the spatial response. Duke's LIAR test would also account for this. The "next room" effect -

Gary

 

[ahofer]

“Cannot” be measured is too strong a claim. That measuring these things is difficult or highly complex, or that we don’t have the right suite of measurements, I would readily accept.

 

[geickmei]

“Cannot” be measured is too strong a claim. That measuring these things is difficult or highly complex, or that we don’t have the right suite of measurements, I would readily accept.

You might be able to correlate some measurements to perception but you can't replace perception.

Gary

 

[ahofer]

You might be able to correlate some measurements to perception but you can't replace perception.

Gary

Oh I agree you can’t measure things that aren’t in the signal(s) by measuring the signal(s).

I suppose there are some brainwaves and such that would offer a clue, but that’s out of scope. Besides, that sort of stuff isn’t limited to speakers.