Klipsch R-41M Bookshelf Speaker Review

[tuga]

John Atkinson uses pink noise too.

But I am still unconvinced that it is useful for anything other than assessing tonal balance.

 

[Dimitri]

I don't think that's it - it's that trained listeners are much more attuned to listen to flaws. So naturally if they hear a flaw, they will be pickier about what constitut a truly great speaker. I took Harman's freely available listening course(on mobile, so can't find the link right now) and that alone was revealing for helping figure out flaws. Just because I can tell flaws in speaker more easily doesn't mean I enjoy them less though.

I downloaded it, but right now I have to run out to catch the sunset in Ventura... <arnold> I'll be back </arnold>

 

[amirm]

With some groups being excited and other being "meeh" I don't see how we can say the voted the same.

They voted the same with respect to which speaker sounded better to them than another. As explained, trained listeners are much more critical as they should be. Young listeners were found to be the opposite, not being bothered a ton by even the worst speaker.

 

[Francis Vaughan]

John Atkinson uses pink noise too.

But I am still unconvinced that it is useful for anything other than assessing tonal balance.

It will reveal other nasties. Resonances etc.
In principle it is providing equal energy per octave right across the spectrum, so it just acts to energise everything at once. Whether you want to call this the agregate tonal balance or not is maybe a matter of semantics. In the end, tonal balance is what the Klippel analysis is delivering. Frequency response as heard. It just does it with a lot more rigour.

 

[tuga]

It will reveal other nasties. Resonances etc.
In principle it is providing equal energy per octave right across the spectrum, so it just acts to energise everything at once. Whether you want to call this the agregate tonal balance or not is maybe a matter of semantics. In the end, tonal balance is what the Klippel analysis is delivering. Frequency response as heard. It just does it with a lot more rigour.

I need evidence to prove that potential problems resulting from the reproduction of sharp transients are revealed / not masked by steady state full range sound.

 

[Hugo9000]

"Most audiophiles" have been tested and their preferences very much match trained listeners. Here is a formal test of trained listeners versus many other groups of population including audio reviewers:

View attachment 51283

As you see, the ranking of the speakers (each colored graph) stays the same no matter which group is listening.

I have taken the test twice myself and both times I voted the same as the majority/trained listeners.

You just need to put aside your intuition on this front and go with science and research.

Of course this is for blind tests. Make it sighted and many factors impact preference from story behind the speaker, to its cost, shape, etc.

It's interesting that the reviewers don't place B&W too far below the highest rated speakers, while nearly all other groups including trained listeners have quite a large dropoff in how they rate B&W compared to the top two scorers. Could it be that the reviewers have become 'acclimated'/'burnt-in' to the "B&W house sound," and thus are less bothered by it than most other listeners (except group Acad7)? It seems likely that many/most seasoned/experienced reviewers would be quite familiar with the famous models from B&W since they've received so much coverage over the years from the magazines, and I wonder how many reviewers have personally owned them or had them on 'long term loan.' lol

Of course it could be coincidence, or the reviewers and Acad7 have lived with speakers with similar characteristics to the B&W. I think it's been revealed that "B" stands for Bowers & Wilkins, but I can't recall which specific model was tested.

I could be imagining a pattern where none exists haha. The best I can say from my own personal listening experience with B&W is that I wanted to like them. (Actually, I rather expected to like them, based upon reading reviews and seeing them listed as monitoring speakers for so many of my favorite Deutsche Grammophon recordings lol I was more than a little disappointed.)

 

[tuga]

Perhaps a CSD plot can illustrate my point. One could argue that by looking at a FR plot we can determine that resonances exist and even speculate with some accuracy the frequency of those resonances. But that's about it.

source: https://audioxpress.com/article/testing-loudspeakers-which-measurements-matter-part-1

 

[tuga]

It's interesting that the reviewers don't place B&W too far below the highest rated speakers, while nearly all other groups including trained listeners have quite a large dropoff in how they rate B&W compared to the top two scorers. Could it be that the reviewers have become 'acclimated'/'burnt-in' to the "B&W house sound," and thus are less bothered by it than most other listeners (except group Acad7)? It seems likely that many/most seasoned/experienced reviewers would be quite familiar with the famous models from B&W since they've received so much coverage over the years from the magazines, and I wonder how many reviewers have personally owned them or had them on 'long term loan.' lol

Of course it could be coincidence, or the reviewers and Acad7 have lived with speakers with similar characteristics to the B&W. I think it's been revealed that "B" stands for Bowers & Wilkins, but I can't recall which specific model was tested.

I could be imagining a pattern where none exists haha. The best I can say from my own personal listening experience with B&W is that I wanted to like them. (Actually, I rather expected to like them, based upon reading reviews and seeing them listed as monitoring speakers for so many of my favorite Deutsche Grammophon recordings lol I was more than a little disappointed.)

The study is about preference. And to use a beaten cliché, "there's no accounting for taste"...even that of "professional" reviewers.

 

[Francis Vaughan]

I need evidence to prove that potential problems resulting from the reproduction of sharp transients are revealed / not masked by steady state full range sound.

Can you charaterise the physical nature of such a problem? Sure, pink noise analysis won't find things like dynamic compression. But if you are looking for things that appear in an impulse response, you are going to find them in pink noise. There is a mathematical duality between the two. There is a lot of handwringing about transient response, but in the end it is mathematically identical to frequency response, unless it specifically involves things like thermal compression, it is just impulse response, which is the dual of frequency response. Thermal compression effects are long terms effects on the time scale of a single impulse.

 

[wwenze]

"Most audiophiles" have been tested and their preferences very much match trained listeners. Here is a formal test of trained listeners versus many other groups of population including audio reviewers:

View attachment 51283

As you see, the ranking of the speakers (each colored graph) stays the same no matter which group is listening.

I have taken the test twice myself and both times I voted the same as the majority/trained listeners.

You just need to put aside your intuition on this front and go with science and research.

Of course this is for blind tests. Make it sighted and many factors impact preference from story behind the speaker, to its cost, shape, etc.

Now the question is, which speaker is flat and which speaker is the Klipsch

 

[Jukka]

Whats the point of all this with commercial speaker of such a low cost? You can't even begin to have a good speaker for that money, unless you DIY. At least compare it to other speakers of the same price point.

 

[amirm]

Whats the point of all this with commercial speaker of such a low cost? You can't even begin to have a good speaker for that money, unless you DIY. At least compare it to other speakers of the same price point.

I compared it to the Pioneer which is close in price.

As to your complaint, we are building a library of measurements and it is critical that we have a range of performances in there. And there are plenty of situations where you want secondary or tertiary speakers (garage, etc.). Besides, I am the one doing all the work. What are you complaining about?

 

[QMuse]

It will reveal other nasties. Resonances etc.
In principle it is providing equal energy per octave right across the spectrum, so it just acts to energise everything at once. Whether you want to call this the agregate tonal balance or not is maybe a matter of semantics. In the end, tonal balance is what the Klippel analysis is delivering. Frequency response as heard. It just does it with a lot more rigour.

All peaks in FR are coming from some kind of resonances hence they all fall into "tonal balance" cathegory.

I need evidence to prove that potential problems resulting from the reproduction of sharp transients are revealed / not masked by steady state full range sound.

Reproduction of sharp transients found in music (like cymbals, etc) is also about even FR (tonal balance). People just use different words when describing disbalance ("harsh", "snappy", etc).

 

[pma]

But if you are looking for things that appear in an impulse response, you are going to find them in pink noise. There is a mathematical duality between the two. There is a lot of handwringing about transient response, but in the end it is mathematically identical to frequency response, .

Impulse response is exactly dual to frequency response only in case that you speak about complex frequency response (or better say complex transfer function) which includes both real and imaginary part or amplitude and phase response. As "frequency response" here it is usually meant amplitude frequency response with no regard to phase. Then, the mere amplitude response is not enough to describe transfer function of the system. And we are still restricted to linear systems.

Same amplitude spectrum and missing phase information may lead to completely different waveforms in time domain.

All sine sources in phase

Effect of different phases of sine sources

 

[QMuse]

Perhaps a CSD plot can illustrate my point. One could argue that by looking at a FR plot we can determine that resonances exist and even speculate with some accuracy the frequency of those resonances. But that's about it.

source: https://audioxpress.com/article/testing-loudspeakers-which-measurements-matter-part-1

It has been said many times - resonances are either visible in FR or if they aren't then why care about them. Now, if they are visible/audible in FR I don't really see why should I care if peak in FR is cocming from cabinet resonance, driver resonance or some other resonance - there is nothing I can do about it anyhow. I really don't understand why anybody besides speaker designers/DIY builders should look into CSD when everything that needs to be shown is already there in spinorama graphs as that is where it all sums up.

 

[Jukka]

Besides, I am the one doing all the work. What are you complaining about?

That time and money is away from the interesting speakers.
Okay, I get the library thing. My suggestion: get as many different constructions as you can: 2-way, 3-way 4-way, classic rectangular boxes, wide baffle, narrow baffle, wave guides, horns etc. I'm really interested in the directivity patterns at play.

 

[pma]

Speaker Listening Tests
On my right I had the Pioneer SP-BS22-LR and on the left the Klipsch R-41M. With levels matched and playing one channel at a time,

I can hardly imagine more unscientific method of listening test comparison, as to have one speaker under test on the left and the other one on the right. Regardless you used level matching and same signal.

 

[Sancus]

I can hardly imagine more unscientific method of listening test comparison, as to have one speaker under test on the left and the other one on the right. Regardless you used level matching and same signal.

Hard to imagine how that makes it any less "scientific" than doing it sighted to begin with?

That time and money is away from the interesting speakers.

It's important to remember that the higher the cost of the speaker, the fewer people are interested in it for simple reasons of economics. Testing a bunch of budget speakers puts the data in context of what can be accomplished at all price points, and that's important. Not to mention that cheap, small speakers are both easy to obtain and easy to set up on the NFS.

Besides, there are expensive, high performance speakers being tested as well, like the Neumann KH-80 and upcoming shortly Genelec 8341a.

 

[amirm]

I can hardly imagine more unscientific method of listening test comparison, as to have one speaker under test on the left and the other one on the right. Regardless you used level matching and same signal.

Again, that is because you have no experience in this field. Symmetric placement of AB speakers is poor man's version of putting both speakers in the same spot which requires a mechanical shuffler.

Instead of constantly complaining, maybe you outline how you are testing speakers according to whatever science you are imagining.

 

[tuga]

Can you charaterise the physical nature of such a problem? Sure, pink noise analysis won't find things like dynamic compression. But if you are looking for things that appear in an impulse response, you are going to find them in pink noise. There is a mathematical duality between the two. There is a lot of handwringing about transient response, but in the end it is mathematically identical to frequency response, unless it specifically involves things like thermal compression, it is just impulse response, which is the dual of frequency response. Thermal compression effects are long terms effects on the time scale of a single impulse.

You are trying to convince me that by listening to a single speaker playing pink noise (steady state sound) you are able to hear a driver resonance that doesn't produce an obvious peak in the FR response or that the bass reflex alighnment is uderdamped, or that a midbass cone is not particularly good at reproducing low-level detail.