Should we (I) get into speaker testing & measurement

[pwjazz]

some manufacturers are going to their labs and making measurements there

I assume their lab is in Europe? Perhaps we could structure this as setting up a "west coast" Klippel lab and help fund the venture by renting time to manufacturers and other interested parties to get lab time? The best thing to do with expensive equipment like this is to make sure it doesn't sit idle too much, otherwise it's basically just wasting money.

 

[daftcombo]

Hi @amirm ,

I would be very interested in your speaker measurements. That would break new grounds in hi-fi.

Will you also go into disassembling them, as this guy do before measuring?
http://noaudiophile.com/reviews.php

 

[boXem]

I voted yes. It's an ambitious project, thus I like it. Founding it will be interesting.

We are many members from Europe. Would it be an idea to directly deal with Klippel labs in Germany and making this project a community effort?

I am just wondering if speakers are not slowly becoming a thing from the past. This idea doesn't fit well with heavy investments.

Last but not least, several members believe that headphones should have higher priority. I would tend to agree with them, especially since the investment would be more reasonable.

 

[MZKM]

I am concurrently researching headphone testing with this project. It has its own set of issues. Cost for a high-end system is lower but we are still talking $10K to $20K. These guys charge a fortune for everything.

I will probably start a new thread on headphone testing and get feedback on interest and issues involved. From logistics point of view, headphones are of course much easier to manage. The Klippel system at the extreme requires 14 foot high ceilings! I only have one spot in the entire house for it and is definitely not ideal.

I feel Rtings pretty much has headphone (and tv/monitor) testing in the bag. And they even had Sean Olive come over to look over their testing methodology. The only thing they don’t currently do (they said they are planning to) is impedance/phase and driver sensitivity, like how InnerFidelity’s measurements would tell you how much wattage/volts you need for 90dB. I wish they’d subjectively scored cable microphonics as well, as that can be annoying.

 

[pwjazz]

Would it be an idea to directly deal with Klippel labs in Germany and making this project a community effort

I like your thinking!

 

[pozz]

This is very exciting.

 

[Kal Rubinson]

A proper multi room system, to have music everywhere in my house, with good streaming service (Qobuz in my case), streamer software (Roon?), streamer hardware (any, actually), DAC (done, with different levels of cost, comfort and quality), Power amp (done, but only at a cost) and room correction (well, what's my choice outside of Dirac? But that limits my hardware choice or increases the costs too much. So, not 'done', for me)

DiracLive can be run on a PC/Mac streamer, the same one that is running Roon (or JRiver), so there is no real limitation on hardware choice.

 

[617]

Amir, let me provide a few comments. I am a diy speaker guy and I've measured a few speakers in my day. I do them indoors using the quasi anechoic technique used by many hobbyists, Stereophile and so on.

First, you are completely correct that these indoor measurement techniques do have advantages over even anechoic measurement. This may be a difficult myth to dispel but it is important. If you know what you are doing, and integrate the near field and far field measurements, you have a good picture of frequency response. Also, regarding spurious noises - this is not actually a problem. You generally send a short measurement signal (I use MLS signals) of about 4 seconds to the speaker - then repeat 3 times. It is very easy for software to identify any spurious noises by comparing the 3 measurements.

Second, the Klippel measurement system is amazing stuff, precisely because it provides a very 3d view of the speaker. Most of the time, people measure horizontal directivity and vertical as well, but I think that a full sphere measurement provides a better view. Earl Geddes has been sharing his technique/code for a modal measurement technique on DiyAudio. I am not totally familiar with his measurement protocol, but basically he uses fewer measurements behind the speaker, where there is less treble, and more measurements in front. I can put you in touch with the people who are continuing this work. You seem to have experience with programming and audio, so you may be the ideal person to pick up Geddes' work.

Geddes uses a simple rotating platform to do his measurements.

If you want to invest in a good measurement setup, I would recommend buying or building a turntable which works with ARTA or some other measurement package. These are not nearly as advanced as the Klippel products but will provide an accurate view of speaker directivity.

https://outline.it/outline-products/measurement-systems/et-250-3d/
See this measurement platform. I believe ARTA works with this one out of the box.

Now, regarding the utility of measurements. Harman's conception of the ideal speaker probably does reflect most listener preferences, but it seems to me that if you want to really interrogate speaker performance (and value) there are other speaker measurements which need to be done. You mention that bass performance is an important predictor of listener preference, but bass performance can't be completely characterized by FR measurement, especially with small speakers. It's essential to know how the speakers sound when they play loud! This requires distortion measurement, which is difficult to do in a repeatable and comparable fashion, but the equipment involved is simple - calibrated mic capable of high SPL.

Please see these ARTA application notes: http://artalabs.hr/support.htm
Especially:
AP4 - Loudspeaker free field response
AP5 - Chamber for LF mic callibration
AP6 - Directivity measurements
AP7 - Displacement estimation using STEPS (ARTA's distortion analysis component)

 

[vert]

I voted yes, and would be able to contribute a modest amount if the project comes to light. ASR provides an amazing, necessary service. I have a recent anecdote. A rather wealthy, older friend of mine is into audio. One day he mentioned to me excitedly about a purchase he was contemplating. He had been to a one of the fancy audio salons we have here in Switzerland and explained to the salesman how he would like his speakers to sound more deeper and more spatious (his wife restricts him to bookshelf speakers). The salesman recommended a $ 5000 power conditioner to achieve that, which my friend was offered to take home on trial. He was pondering whether or not to buy the thing. He was a little worried he might get a slap on the wrist from his wife. He had no doubt a power conditioner would spectacularly improve his system. I pointed him to a couple of ASR threads on power conditioners and on room correction; I don't know what he did eventually. The whole incident made me realize the power of persuasion the audio industry has over people who like my friend, are very intelligent and have distinguished careers in their fields, and that lack of objectivity and testing have to be a big part of the problem.

 

[Thomas savage]

It's a great idea, you need a assistant imo and if the community want it they should fund it .

If they can't fund it I'd leave it for now.

Not keen on sponsorship, we should remain funded by the membership and totally independent. Nothing is worth comprising that.

 

[Juhazi]

I just voted No. It would require enormous funding and facilities and easily flop. A wonderful idea, but...

 

[Krunok]

This is how RTings does it:

How do we make money and how can you support us?
We don't have ads on our website, because we want to stay unbiased (for example, some other review websites have ads from Samsung). Instead, we only make money in two ways:

1. From contributions directly from you here, which gives you access to extra features.
2. If you buy something that we recommend via links to retailers on our website (like this one). That way, we are aligned with the same goals as you: to find the best product for your needs. Learn more about these links.

Where do the products that we review come from?
Unlike most websites, we do not get our review units directly from the manufacturers. Instead, we buy them on Amazon.com and keep them until they are discontinued, and then resell them used locally. It means our units aren't handpicked by the manufacturers, but actually represent what you would buy yourself.

 

[q3cpma]

Yes, obviously. Everything between the recording and the speakers/headphones is mostly solved as a problem, being now just an exercise in implementation.
Headphones don't lack good measurements (diyaudioheaven, innerfidelity, rtings, goldenears, etc...), but for loudspeakers, I only know of Sound & Recording (german and not free) and some trusty manufacturers (Genelec and Neumann, mainly).
As another guy said, doing some (non destructive) dissassembly and basic analysis of the transducers used and the amp/DSP module (for active speakers) would be a nice plus.

 

[617]

Related question - is there a reliable way to measure headphones? Could someone give me a rundown? The ear/transducer interface seems incredibly variable, and the measurements I've seen of headphones suggest really poor performance.

 

[Blumlein 88]

Related question - is there a reliable way to measure headphones? Could someone give me a rundown? The ear/transducer interface seems incredibly variable, and the measurements I've seen of headphones suggest really poor performance.

I have the same concern. The work by NRC in Canada and by Harman convince me they have laid important groundwork. Headphones.....not so much. Listening to Harman based speakers also is convincing. Again with headphones listening is almost a crap shoot. What works for me might(probably) won't work best for you.

 

[SIY]

I have the same concern. The work by NRC in Canada and by Harman convince me they have laid important groundwork. Headphones.....not so much. Listening to Harman based speakers also is convincing. Again with headphones listening is almost a crap shoot. What works for me might(probably) won't work best for you.

Having measured headphones using several different jigs, including the high priced spread, I have come to the uncomfortable conclusion that it's a fool's game.

 

[617]

I have the same concern. The work by NRC in Canada and by Harman convince me they have laid important groundwork. Headphones.....not so much. Listening to Harman based speakers also is convincing. Again with headphones listening is almost a crap shoot. What works for me might(probably) won't work best for you.

B88, it seems like the convention is to use some kind of measurement mic embedded in a plastic ear. Perhaps comparative measurements are possible but it's difficult to create a reliable reference. I would be more interested in seeing ASR try to develop the most useful headphone measurement methodology. Any thoughts on this?

 

[617]

Having measured headphones using several different jigs, including the high priced spread, I have come to the uncomfortable conclusion that it's a fool's game.

This is very bothersome to me. Could you discuss the frequency bands which seem most unpredictable?

I listen to a pair of Sennheiser hd380 pro sealed headphones. They have a pretty smooth FR to my ear, with a good balance between treble and bass, yet the measurements looks like this:
https://www.innerfidelity.com/images/SennheiserHD380Pro.pdf

15db swing from bass to treble.

Part of me just can't believe headphones can't be characterized better. They don't even need to be measured at different points in space!

 

[svart-hvitt]

B88, it seems like the convention is to use some kind of measurement mic embedded in a plastic ear. Perhaps comparative measurements are possible but it's difficult to create a reliable reference. I would be more interested in seeing ASR try to develop the most useful headphone measurement methodology. Any thoughts on this?

You must measure two things:

(1) Headphone freq. resp.
(2) Your HRTF

ASR cannot do (2).

 

[SIY]

Part of me just can't believe headphones can't be characterized better. They don't even need to be measured at different points in space!

The problem is that the response is inextricably tied to the shape of the pinnae (and no two are the same) and the way the headphones fit and compress the ears.

If I even do something as simple as taking the headphones off the jig, then replacing them, I see 5 dB or more frequency response swings. By contrast, when I repeated measurements on some speakers taken in my basement in Illinois before I moved and in my lab in Phoenix after I moved, they were almost identical.

Now, that said, if I see a frequency response feature that persists even when using very different test jigs, I can attribute that to the headphones. But unlike speakers, I can't correlate all of the other parts of the measurement to the sound.