How to "listen"?

[Don Richard]

Pistons moving air on one axis can not create the sonic wave patterns that are created by a strummed guitar (or Mon Dieu, a piano!) in a typical listening room.

One is not trying to recreate the sound of the instrument alone, which would include the instrument's polar pattern. In the case of a recording made with 2 mics, the recordist is trying to capture the performance of the musician playing his instrument in a certain location. If the recordist is skilled with mic selection and placement, the recording will capture the reflections and phase differences that define that instrument's sound in that particular space. In that case the loudspeaker's polar response is irrelevant as long as the listener is within the polar pattern of the loudspeaker and as long as the room in which the loudspeaker is located does not impose it's own sonic signature on the reproduced sound. In fact it is preferable to use a loudspeaker with a tight polar pattern so that the acoustic energy does not hit the walls and cause early reflections which can muddy up the sound. Tom Danley has written about this and he uses his SH-50 Synergy horn in his home stereo. This horn exhibits pattern control over a wide band of frequencies and is a true point source.

It is also wise to treat the room to minimize it's influence on the loudspeaker's reproduction of sound. Here is a good thread on the subject:

https://community.klipsch.com/index.php?/topic/131163-corner-horn-imaging-faq/?hl=%2Bhorn+%2Bimaging+%2Bfaqs

[Mallette]

I think "accuracy" started making big strides about the time they took away our bean bag chairs.

Good post, Dean. Accuracy is an excellent objective IF THAT IS WHAT YOU WANT. It's not a question of right or wrong, but of being happy or not. To some extent, one can separate the accuracy crowd from the "I like it" crowd in that it's the latter who cover their ears at horns. Mercilessly transparent.

But I think it excessive when any audiophile turns his nose up at another's system because he finds it either too accurate (clinical they might call it) or too warm and musical (inaccurate he might call it).

I like both kinds. While it isn't working at the moment, I have a big 1936 RCA console radio with a 12" speaker. Not even a little bit "accurate" but warm, mellow, and comforting to listen to. I like it ALL.

Dave

[Mallette]

Unless the microphones are placed to simulate (represent) a human being, the result is 100% arbitrary.

No competent engineer attempting to accurately record an acoustic space/time event would do it any other way.

If a person is just after some "pleasing" results or other arbitrary goal they can do it however they wish. Certainly that is precisely how engineered recordings, like Sgt. Pepper's, Pink Floyd, etc are all about. "True to the original performance" is absurd in that context, as there was no "original performance."

No value judgments of any kind in the above, just some context.

Dave

[Chris A]

Accuracy is an excellent objective IF THAT IS WHAT YOU WANT. It's not a question of right or wrong, but of being happy or not...But I think it excessive...

I recommend reading this excellent article in its entirety and considering--carefully--what the author is measuring and hearing:

http://www.stereophile.com/reference/406howard/index.html

There is also an excellent article by Nelson Pass on this very subject for amplifiers that refers to "the elephant in the room": cascaded harmonic distortion in real music reproduction, i.e., both low order and higher order harmonics, mixing from more than one instrument playing at once that create opaque-sounding output and distortion that turns non-harmonic.

Chris

Edited January 18, 2014 by Cask05

[Mallette]

I hold in my hand the LP "Clouds" by Joni Mitchell. How do I know where the microphones were placed and what the room(s) was for the various musicians?

I've said the same myself. It's all a guess unless you either made the recording, were present at the event, or have been present at a similar event.

That said, one can get a sense of say, St. Mark's in Venice, from images and such. If you know the placement of the vocalists or instruments as is the case with the LP notes from the great Biggs and company recording made there, you can get a pretty decent feel for "accuracy."

Accuracy is the domain of the engineer. How do we know Piranesi's drawings were "accurate?" We have only our own senses to tell us.

Dave

[Don Richard]

No competent engineer attempting to accurately record an acoustic space/time event would do it any other way.

That contains a lot of qualifiers, a ton of assumptions. Let's make it more real and specific. I hold in my hand the LP "Clouds" by Joni Mitchell. How do I know where the microphones were placed and what the room(s) was for the various musicians? This is not a trick question, nor a rhetorical question. It is the heart of the problem in playing commercial multitrack recordings.

FIFY

You don't have to know any of these things anyway, all you need to do is play the recording and enjoy it. Or not, if you don't like it. What Dave said is correct for acoustically recorded performances.

Edited January 18, 2014 by Don Richard

[Mallette]

[quote name=mdeneen' timestamp='1390067144' post='1683931] To the extent you can transpose and scale measurements from the Pirenesi drawing to the subject drawn, that's the extent you can speak of accuracy.

That's the reason I used him. Many of the things he drew meticulously are now gone. Others remain, though changed. Do we assume all are accurate because we can verify that some are? Perhaps not for courtroom standards, but I am good with it.

One may use the same rubric to judge the accuracy of a recording. If you know the venue and it's acoustic properties well, even an inexperienced listener is likely to be able to guess the number and location of microphones in a recording of the same place. Human hearing is considered by many to be our most highly develop sense.

Anything done from more than a single point in a space starts off inaccurate. Not that accuracy cannot be re-created in the hands of a skilled engineer.

A brief anecdote on that line. I've spoken of George Ellis Mims, a friend I consider one of the world's great living musicians. He was organist and music director at the great St. Martin's church in Houston for decades. He knew every nuance of the building and instrument. He used the chief audio engineer of the Houston NPR station for recording. This guy used multiple microphones and such in 180 degrees to my own "two mikes, best seat in the house" approach. I had the privilege of sitting in on one of the final edit sessions with George and Todd at the station for George's "swan song" organ CD. I was blown away at the guys skill. I still think he went to a lot of extra effort compared to my own method to achieve essentially the same goal, but when we brought the edit master to my listening room and played it George said "That sounds like the church." George is merciless about poor recordings. I agreed with his assessment.

I remain firm in my opinion that objective accuracy is both possible and is responsibility of the engineer, and that use of that term for anything other than pure acoustic space/time events is by nature subjective.

I would also point out I have recorded in venues where accuracy was the LAST thing you want due to the nature of the acoustics. A stage with a glass back wall at Dallas Fair Park and a converted fire station come to mind...

Dave

[Mallette]

Ringing a bell in my living room?

As I have never heard Clouds I cannot comment on it. As to your statement, yes. That is a pure acoustic space/time event. It can be recorded accurately from a single point by a decent engineer and will reproduce that event from that perspective limited only by the playback equipment. With accurate equipment properly set up in your room, you could have a "ghost" image as the one I reproduced in my SoundCube experiments where everyone who heard it looked towards the front door when it opened in the recording, then appeared confused when it was quite obviously closed.

That is accuracy. It is not complex to understand...but there is considerable distance between understanding and accomplishing it. Given your experience in designing electronics you should be able to relate to that. I fully understand the issues, but I haven't a clue as to how to build an accurate...or even awful...preamplifier. But I know a good one when I hear it.

Dave

[Don Richard]

If something is possible, then you can explain the path by which it happens, or the precedent by which it has already happened. So, when I grab my "Clouds" recording, how do I access this information to make a judgment about accuracy. This is not an arcane question. If I grabbed a 2 x 4 at Lowes and they said it was 94" my path of judging the accuracy would be to use a ruler and measure the 2 x 4. Nothing fuzzy here. If the engineer had some responsibility, how do I access that for "Clouds?"

By having experienced a performance by the artist you have an idea of what it is supposed to sound like. Most likely "Clouds" was done in several sessions, each session adding effects perhaps, or correcting a bad note she hit on her guitar, or punching a correction regarding vocal phrasing or the like. These techniques are done so that the recording sounds as perfect as possible, and that goal was probably reached with this particular recording. There is no value in knowing exactly how the recording was made. One only needs to know that the artist was satisfied with the results and that you are listening to a product that shows the artist in the best possible light. If listening to this recording transports one to the place and time that one heard the performer live, then it is accurate.

There is a whole bunch of BS going on in the audio world related to the inability to re-hear a sound that occurred in the past. What this results in is people basically writing crap like mdeneen has written, likening music to a 2x4. Trained listeners can easily remember sounds, remember the melodies, remember the lyrics, and recreate those works if they have the talent to do so. Those who have no such talent are reduced to turning to epistemological arguments. It's all they know how to do.

[Mallette]

Dam, DR. Now I remember why it's not so good to be on the receiving end of your fire...

Dave

[Mallette]

Just to keep my head down, accuracy is NOT a value judgment. I love great engineered recordings. Accuracy isn't relevant to them. "Dark Side of the Moon" in a great pressing or digital media is a totally wonderful experience on a fine system. It puts one in a place and space that is entirely engineered. In my case, it's one of the places where my DynaQuad setup shines and sounds swirl around the room. Accurate? Heck, no, thank God. Just WONDERFUL.

Not only nothing wrong with that but the world of music would be terribly poorer without it.

Dave

Edited January 18, 2014 by Mallette

[Mallette]

What kind of judgment is it then?

If I was not clear I'll wait to see if anyone else is confused concerning actual acoustic space/time events as opposed to created or augmented ones.

I am quite satisfied with the clarity of my statements.

Dave

Edited January 18, 2014 by Mallette

[Don Richard]

There is a whole bunch of BS going on in the audio world related to the inability to re-hear a sound that occurred in the past. What this results in is people basically writing crap like mdeneen has written, likening music to a 2x4. Trained listeners can easily remember sounds, remember the melodies, remember the lyrics, and recreate those works if they have the talent to do so. Those who have no such talent are reduced to turning to epistemological arguments. It's all they know how to do.

No, it's not epistemology at all. It's elementary logic and nothing else. Accuracy is what is under discussion. That's an objective measure of something to something else. We have a hifi system of some kind, and a pile of records and we are asking is this system accurate when playing them back. The answer to that question is to compare them to a standard and see how alike or different they are from the standard.

Let's examine your first claim there. You want to compare the playback to something you heard in the past, if I have distilled your first sentence properly. It means when I playback Patti Page's Greatest Hits (my sample A), I derive my sample B by re-hearing sounds from the past. Which sounds? I have never been to a Patti Page concert, she hasn't been in my room for me to listen to. How about I compare her to Judy Collins whom I have heard? Will that work for an accuracy measure? I heard Debbie Reynolds once in Vegas, will that be close enough? I was sitting way in the back (cheap seats you know) will that be the same as sitting 10 feet from the speakers?

Please notice that I am not asking questions about the structure and meaning of knowledge. I am asking simple direct questions about the problem at hand. Question to answer the logical problem of how to make an A to B comparison for the sake of knowing something about accuracy. So, logically, your first claim just doesn't work. It works in highly controlled circumstances like recording a door closing and then playing back. But that's not the essence of what we do with stacks and stacks of CD/LPs, is it? I've never heard Maria Callas at the Met. I've never heard Miles Davis in a studio.

Wow. So, according to you no recording can possibly be accurate? Imagine that you are at a football game and the quarterback throws a 50 yard pass that the receiver runs in for a touchdown and you see this happen. Then the Jumbotron shows the replay. According to you this replay cannot be accurate, even though you saw the TD pass and what you see on the video screen depicts exactly what happened. According to you the actual play happened in the past as a fleeting event and thus cannot be directly compared with the video to determine the video's accuracy so it is inaccurate. What an absurd notion. And if I watch a film of Y. A. Tittle throwing a TD pass it is not an accurate depiction because I was not yet born when it was filmed so I have no reference to determine if it actually happened? That's why we record things, to preserve them for the future. Recordings can be a historical record of an actual event.

Edited January 19, 2014 by Don Richard

[garyrc]

In my early era, Edgar Villchur put on a big demonstration of showing the AR-3 was the most accurate speaker because people in an auditorium couldn't tell when his speakers started and the chamber orchestra stopped. It was nice showmanship, and for anyone who had a living room the size of an auditorium (including stage) it was really meaningful.

Klipsch did essentially the same thing, and included it in an ad in HiFi/Srereo review in 1967: "A special concert was staged in which Klipschorns reproduced, at original loudness, the Hartford Symphony Orchestra. This was a live v.s, recorded concert, and the majority of the large listening audience could not tell the difference between the live orchestra and the sound of the orchestra as reproduced by Klipschorns."

Yet, Klipschorns and AR-3s sound very different to my ears.

PWK might point out that this was a full orchestra "at original loudness," as opposed to a chamber orchestra.

The large rooms used in both demos do, of course, differ from most home listening rooms.

Another demo, with Khorns v.s. a small jazz group, was done at Berkeley Custom Electronics in a room that was probably close to the size of our home venues (3,000 cu ft?). The results were similar.

I wonder how any or all of these tests would have come out with an audience of today's audiophiles.

Edited January 19, 2014 by Garyrc

[Deang]

Good stuff by everyone.

I'm stuck on Chris' "pipe organ" post (enjoyable read btw).

It's as if we need to separate the tones from the effects that come with higher SPLs.

It's absolutely true - if you want the live level experience, you're going to need the big stuff. However, what if you don't require this effect to derive pleasure from your listening experience? What if instead, you find that you're the kind of person that develops an emotional connection with the music and receives the highest level of satisfaction in a completely different context?

I stopped engaging in live level listening around the time I was contemplating selling my Klipschorns. Many might remember me tearing down my AK-4 Klipschorns and building false corners, installing new horns, and networks. The motivation was in part driven by the realization that after buying them, the only thing they did well was blowing me out of the room. They sounded awesome at 90dB and above, but at "normal" levels, I thought they sounded dark, recessed/flat, and in a word - boring. It was impossible to have an intimate listening session with them.

I started chasing something I'd lost, and it didn't have a damn thing to do with bombastic pressure waves - which quite frankly, I was very tired of.

[Mallette]

When I play two channel then switch to 6 channel mode, the sound becomes much more natural and effortless - up to levels that are closer to live concert levels for these acoustic ensembles. I can feel those experiences again--at least in part.

While I am quoting your mention of multi-channel because of my own constant, if unpopular, efforts to point out that anything like "fidelity" is impossible with 2 channels only, I love your whole post.

Those of use exposed to and intimately familiar with pipe organs learn some things others have a hard time with. Of all instruments, the organ presents the greatest challenge to the engineer in really capturing anything like an accurate image. Then, it is also the greatest challenge to a system to reproduce.

The space is totally critical. While one can have a very satisfying experience with a well made 2 channel recording, it cannot have anything like real fidelity as the space that organ fills...especially if it has divisions to the rear or other places...is absolutely critical to the listening experience. The four organs of Freiburg played simultaneously would be an extreme example. Then, there is the dynamic range. I've never measured it, but I think a good guess is that the Schoenberg instrument in St. Martin's of Houston probably has a range from well below 30db to well over a 100 with it's double expression swell architecture. Further, the 32' Leiblich Gedeckt can sound very, very quietly...but at 16.5 Hz that takes a very efficient speaker to reproduce with anything like the real subtleties of that sound. IMHO, only a large, horn loaded speaker need apply.

Another purely "IMHO" is that horn loaded speakers in general do far better with the pipe organ than any other design.

Thanks, Chris. Great stuff.

Dave

[garyrc]

The large rooms used in both demos do, of course, differ from most home listening rooms.

I have seen these demos tried in homes several times. It usually went like this. A guy would play a guitar and sign in Joe's living room and be recorded on a nice quality recorder. Then the tape would be played back on Joe's stereo using Joe's "accurate" speakers. Well, in short, it never sounded like the live guitar. People would laugh, and then offer the usual rational for why it was so far off the real sound.

Interesting you would mention guitar. IMO, good speakers with horn loaded midrange and treble (at the least) seem to sound better, more convincing, more "there," than cone and dome speakers, or any other kind I have heard with acoustic guitar. I haven't heard guitar through ribbon or electrostatics. Guitar sounds "real" with the better Klipsch and JBL and Altec, again IMO. I once knew an acoustic guitar player who had a variety of speakers, but always used the horn loaded ones (Altec) to playback his recordings of himself playing his guitar. I heard one A-B comparison at his house, and agreed.

Edited January 19, 2014 by Garyrc

[Don Richard]

Summary. You are mixing up different "kinds" of events. One is a comparison of A to B to measure accuracy. The other is simply seeing A and asking if it is authentic - did it happen? No measure is being done, because it is only an A and has no B. I say stick to the question of accuracy - in which A and B must be compared to make a judgment.

So when my buddy SMAARTed my Khorns then did a transfer function and the input and the output lined up on a disc that the performer deemed an accurate record of the event, that's not good enough for you? It was an inferred measurement, and most of what we measure are inferred measurements. Knowledge of the coefficient of expansion of mercury, for example, is used in thermometers to determine temperature. Fluid velocity is measured to determine flow, as is differential pressure across an orifice. AFAIK no direct measurement method exists for the determination of real time flow or temp. Did you check your tape measure against a standard before you measured your 2x4? If you didn't then how do you know the measurement was accurate? The best you can say in that case is that the two pieces of wood that you measured were the same within the tolerances of your equipment , and you didn't do that. Not that any of it matters anyway because the dimensions of wood vary according to humidity and the dimensions of anything varies with temperature. If you want to get that deep into "measurement" then nothing can be accurately measured, according to the Heisenberg Uncertainty Principle. You just gotta have faith, man, you gotta have faith.

Edited January 19, 2014 by Don Richard

[Quiet_Hollow]

People do SMAART tests on speakers all the time. What sort of inferences are you trying to make from that test? Are you trying to say something about your speakers versus other speakers?

[tigerwoodKhorns]

I'm slowly getting my 2-channel rig setup and dialed in and like what is being thrown at my ears, but worry about the day one of you comes over and informs me that my system is crap when assessed by universal standards. I read comments by guys on the internet about such and such person taught them how to listen, but never anything further on what that means or how to learn.

Thoughts? Help? Suggested reading?

I know it's ultimately about what sounds good to me, but I don't want better informed guest walking away filled with sympathy and regret for coming out to sample my gear.

8 pages in, did you get your answer?