Do we crave distortion?

[MattHooper]

Hi @MattHooper, intersting measurments, what position is your volume control set at for critical listening? Do you know the input impeadance of your power amps?

Here are my amps:

Conrad-Johnson Premier Twelve monoblock amplifier Measurements

Sidebar 3: Measurements Following its preconditioning test, the Conrad-Johnson Premier Twelve was hot—typical for a tube amplifier. The bias was found to be slightly off during the distortion measurements, was recalibrated, and the prior measurements were rechecked to ensure that the results had...

www.stereophile.com

I'd say typical volume on my CJ preamp is set between around 35 to 48 (if that tells you anything).

However...that's with the CJ running through the Benchmark Preamp (the Benchmark input is set at unity gain). That bumps up the signal gain hence I can hear a bit of tube pre-amp hiss through the speakers when running through the Benchmark, but not normally with the CJ straight to the amps. And straight to the amps means I usually had to turn the preamp up more for the same volume at the listening seat. (I'm forgetting my usual numbers when listening to the CJ preamp direct in to the amp, because it's been a while).

I did bring up in another thread that at one time I did a comparison of the CJ preamp run directly to the amp, vs through the Benchmark, and it *seemed* to sound a bit better (same volume at seating position) run to the amp, using a higher volume setting. I figured this was dubious, but it has also been pointed out that preamps can have a sweet spot in terms of their distortion characteristics with regard to volume control settings, so I still wonder if I might have heard something related - that without running through the Benchmark my normal CJ volume settings are closer to optimal operation (or just my imagination).

 

[pseudoid]

I did bring up in another thread that at one time I did a comparison of the CJ preamp run directly to the amp, vs through the Benchmark, and it *seemed* to sound a bit better (same volume at seating position) run to the amp, using a higher volume setting. I figured this was dubious, but it has also been pointed out that preamps can have a sweet spot in terms of their distortion characteristics with regard to volume control settings, so I still wonder if I might have heard something related - that without running through the Benchmark my normal CJ volume settings are closer to optimal operation (or just my imagination).

It is probably wrong way to look at what you wrote but... [advance apology]
So you really are listening to the CJ (thru the Benchmark) and not listening to the LP directly. hmmmmm

 

[MattHooper]

It is probably wrong way to look at what you wrote but... [advance apology]
So you really are listening to the CJ (thru the Benchmark) and not listening to the LP directly. hmmmmm

Sorry, could you clarify? What do you mean by not "listening to the LP directly?"

I use both digital and vinyl sources for listening.

The Benchmark LA4 output goes to my amps. Both my Benchmark DAC2L and my Phono stage allow multiple outputs, so I've run the outputs to both my CJ preamp and my Benchmark preamp, so both are receiving the outputs from vinyl and my DAC.

I have put the CJ on one of the LA4 inputs. When I choose the "CJ" input, that means my DAC/Phono stage are run to my CJ preamp, and then out the CJ in to the Benchmark, hence the CJ is "in the loop" and I use the CJ for volume control (Benchmark set to unity gain).

Otherwise, I can switch out of the CJ input to the regular DAC and PHONO inputs in the Benchmark, in which case the sources go directly in to the Benchmark and out to my amp.

Hope that clears things up.

 

[pseudoid]

I had a pristine collection of about 600 LPs.
I truly hated "using" them.
I knew the excruciating pain my LPs felt from the moment the needle was dropped on their surface.
I am not kidding.... I would replace any LP that had any more than a few dozen playings.

The very 'first' drop of the needle did not happen on any new LP, until my cassette recorders were properly calibrated to record that initial playing!
Once the CD format came to play; I immediately started mirroring my LP collection and added another 600+ to my library.
Sold off the LP collection. Dispensed with all of the road-trip cassettes. Then, ended up turning the whole CD library to full digital before selling them off too.
Has anyone ever tried to back-up/archive a digital music library (on a 1GbE NAS) of 4TB? << NOT pretty!

I can't wait for the future of an implantable chip for my music library w/a built-in synaptic media player and a distortion generator.

 

[MattHooper]

pseudoid, I'm still unclear on the point you were making in your previous reply to me (not 'listening directly to LP') . I don't know if I addressed your point or not.

 

[pseudoid]

and it *seemed* to sound a bit better (same volume at seating position) run to the amp, using a higher volume setting. I figured this was dubious, but it has also been pointed out that preamps can have a sweet spot in terms of their distortion characteristics with regard to volume control settings,

Sorry, could you clarify? What do you mean by not "listening to the LP directly?"

These statements seem to endorse the thread title.
Hoping I am not taking them out of context (or cherry-pickin')...
At a minimum, it brings up the question of "compared to what?"... dubious that it would be re:Original.

 

[MattHooper]

These statements seem to endorse the thread title.
Hoping I am not taking them out of context (or cherry-pickin')...
At a minimum, it brings up the question of "compared to what?"... dubious that it would be re:Original.

The CJ run directly to the amplifiers seemed to sound better compared to running the CJ through the Benchmark Preamp (which upped the gain of the signal).

Better in terms of my subjective appraise. It's been a while since I did that comparison, but by "better" I seem to remember it sounded a little cleaner, slightly less fuzzy.

Hope that helps.

Totally anecdotal and I'm not even sure I'd think the same if I tried that again, so it doesn't need to be taken as fact.

 

[Sal1950]

Once you start digitizing your vinyl you've given up.

I'm a survivor.

That's what the alcohoic says about going to AA. LOL

 

[JaMaSt]

Slightly OT, but because I love pre-WWII music of various kinds I've always wanted a plugin that would give my home recordings the sound of a worn 78. I've tried some and not liked them. I've also tried recording the lead in and between the notes parts of worn 78s and applied them to clean recordings without much success. Any suggestions?

You slightly OT?

Hold my beer!

https://www.nexusmods.com/fallout4/mods/18541

Q: What does this mod do?
A: This mod makes it so that music played from the Pip-Boy doesn't sound like it's coming from a modern MP3 player, but instead, an in-game radio.

I looked for this Mod because the in-game radio music sounded too good!

 

[JaMaSt]

I fell in love with Rach's 2nd listening to a (cassette) recording of Katchen's in the '80's. It still carries more "weight" - to me - than another recording I own by Grimaud which is audibly far superior.

 

[Barrelhouse Solly]

So you must have started when you were 2?

So I always have two questions for someone who
plays an instrument. I assume the acoustic guitar is designed to project outward. Does an acoustic guitar sound the same to you when you are playing it as it does when you hear someone else playing it? Does an amplified acoustic guitar, outdoors that you see at many places, atmosphere type music, no vocals, sound like a real guitar to you? Or a real guitar amplified. Not sure I’m asking that right. If it never sounds like the un-amplified natural sound of an acoustic guitar, then a recording never will for you.

What about if you have visuals to go with it, does it make it more like the real thing, less, no different ? Like a video with high quality sound? I find, if the live recording is high quality, the visuals add to the realism vs, same recording audio only.

I think it’s a highly complex question as to why something might never quite get to what something sounds like in person. I have had numerous situations where I had a recording of solo acoustic guitar up stairs in a larger open area, and when people who have come over have remarked “who is that playing the guitar?” In disbelief when told it’s the stereo, and some of those would know what they sound like live, in a residence. I have heard others talk of similar experiences, hearing an instrument and discovered it was coming from speakers somewhere.

The common denominator of people who described being fooled has been the speakers, range of FR and how flat, and the room, physical layout.

For natural, life like guitar, but with vocals, one of my go tos is Muddy Waters Folk Singer. I sure there are many others.

Guitars sound different from the back and from the front. I've listened to my guitars played by others. I don't find that visuals add much to music, at least from the audio standpoint. I enjoy watching people play but when I'm doing pure listening, live or recorded, I like to close my eyes or focus on a wall or a piece of furniture. I don't think I've ever heard a guitar recording in a perfectly tuned room. Like I said, I'm talking about my own experience which is subjective. I started on guitar when I was 15. I washed out on clarinet when I was 7. I learned a couple of scales and then wanted to play by ear.

 

[Curvature]

... people who prefer the distortion that tube amplification ...

I've posted this before.

In almost all cases personal preference is untested. People just say generally that they like this or that based on their limited experience. Very few people check, or have the opportunity to check, if their own opinions hold water for themselves by doing smart listening tests.

To me the question of preference isn't the same as liking or disliking something. It's a comment about something deeper in ourselves, about our biology and cognition. Blinded preference studies are interesting for that reason.

There are many, probably the majority of published papers say the opposite with respect to IMD, it’s audible, and it’s significant in audio perception.

IMD and THD are two measurements, one multitone, one single tone, showing aspects of the same underlying problem. IMD is simply more representative of the kinds of nonlinearities that speakers produce when playing music. Unless you are pushing the speaker beyond its limits, or it's a poorly made speaker, it's unlikely that you will be dissatisfied with the sound because of nonlinear distortion. You are much more likely to notice issues in frequency response. Furthermore, electronics have negligible nonlinearities compared to speakers.

You wrote elsewhere that we don't know why people can tell apart multiple instruments all playing middle C. I know it's generally acknowledged we don't have a complete perceptual model for timbre or pitch, but I would be surprised if your example couldn't be meaningfully challenged and if a practicing psychoacoustician would agree outright. Apart from envelope and spectrum (the most important monoaural factors), you have physical location of each instrument, which involves a phenomenon called binaural unmasking because of differing arrival times of sound at each ear. Then there is also the "venue", i.e., where you are hearing these instruments playing. The ratio of direct sound (coming from the instrument to your ears) to reverberant sound (from the instruments to the walls and then to your ears) also significantly affects clarity. In large rooms like concert halls this is a major problem. That's not a complete answer, of course, but I think it's a good one.

 

[Travis]

IMD and THD are two measurements, one multitone, one single tone, showing aspects of the same underlying problem. IMD is simply more representative of the kinds of nonlinearities that speakers produce when playing music. Unless you are pushing the speaker beyond its limits, or it's a poorly made speaker, it's unlikely that you will be dissatisfied with the sound because of nonlinear distortion. You are much more likely to notice issues in frequency response. Furthermore, electronics have negligible nonlinearities compared to speakers.

I'm sorry, I thought you were referring to distortion in loudspeakers. I'm aware of the difference between IMD and THD. What I was referring to was that there isn't general agreement in the professional literature about IMD as a factor relating to preference. Some say, not much of a factor if any at all (Olive and Toole) others say it's an important factor in the subjective evaluation of loudspeakers (preference), close to or as important as frequency response.

In fact, the AES Technical Section on Loudspeakers and Headphones lists distortion analysis as one of the top areas of investigation for loudspeakers as there is still work to do on this.

 

[Travis]

You wrote elsewhere that we don't know why people can tell apart multiple instruments all playing middle C. I know it's generally acknowledged we don't have a complete perceptual model for timbre or pitch, but I would be surprised if your example couldn't be meaningfully challenged and if a practicing psychoacoustician would agree outright. Apart from envelope and spectrum (the most important monoaural factors), you have physical location of each instrument, which involves a phenomenon called binaural unmasking because of differing arrival times of sound at each ear. Then there is also the "venue", i.e., where you are hearing these instruments playing. The ratio of direct sound (coming from the instrument to your ears) to reverberant sound (from the instruments to the walls and then to your ears) also significantly affects clarity. In large rooms like concert halls this is a major problem. That's not a complete answer, of course, but I think it's a good one.

No, I said "they" don't know exactly why (the people in the white lab coats). They have identified key elements like the "spectral envelope" (which is the harmonics/overtones of the instrument as being the key one (it's listed first in everything I have read) and then several other things like rise time. I have never seen envelope and spectrum, as in two different things, referred to that way in anything I have read, but what I read was probably 30 years ago, so things may have changed.

We know that physical location has nothing to do with timbre perception. I could put you in the middle of the desert, blindfold you, and have 3 musicians play you the exact same perfect note from three different instruments, one at a time, and you would say, number 1 is a piano, 2 is an oboe, and 3 was a clarinet, or whatever the 3 may be. Then we go to your living room, same, then a concert hall, the same, and anybody who has been around musicians playing outdoors, with no walls around, knows this. A guitar on a beach, sounds like a guitar in your living room, a trumpet in the middle of a football field sounds like a trumpet in the auditorium, and 100s of other examples. Location/venue, if it is a factor, isn't much of one. And it's the exact same thing with voice, outside or inside. In the Fall it can be checked out, in football fields all across the US, get three kids in band with 3 difference instruments that can play the exact same note in the same octave (frequency) and get the conductor to turn his back to them and play one at a time and he/she will say, trumpet, saxophone, and whatever else fits in there. flugelhorn maybe.

However, it doesn't matter to us much that "they" don't know exactly how we can tell the difference in the evaluation and perception of audio equipment. Who it is essential to, and why "they" keep trying to develop a working model like with color perception is to help the hearing impaired so they can develop better and better hearing aids primarily with speech intelligibility in mind. Research on this goes on in the US and Europe constantly, where subjective preference is the key. This is where I vaguely recall binaural unmasking being mentioned, speech intelligibility and hearing aids, but not in the context of timbre perception of music. Unmasking to me suggests noise, or frequency???

Here is one of the papers on sound perception in infants we were required to read in the course that was a long, long time ago, if anyone is interested.

 

[Curvature]

I'm sorry, I thought you were referring to distortion in loudspeakers. I'm aware of the difference between IMD and THD. What I was referring to was that there isn't general agreement in the professional literature about IMD as a factor relating to preference. Some say, not much of a factor if any at all (Olive and Toole) others say it's an important factor in the subjective evaluation of loudspeakers (preference), close to or as important as frequency response.

In fact, the AES Technical Section on Loudspeakers and Headphones lists distortion analysis as one of the top areas of investigation for loudspeakers as there is still work to do on this.

I agree nonlinear distortion needs a lot more investigation.

 

[Curvature]

No, I said "they" don't know exactly why (the people in the white lab coats).

When I wrote "we" I group "they" in with us, because we (here) read their work. But that's of course from my perspective.

So, rephrasing, from the latest I know timbre is not a closed topic, but environmental conditions in which listening takes place are still very important.

We know that physical location has nothing to do with timbre perception. I could put you in the middle of the desert, blindfold you, and have 3 musicians play you the exact same perfect note from three different instruments, one at a time, and you would say, number 1 is a piano, 2 is an oboe, and 3 was a clarinet, or whatever the 3 may be. Then we go to your living room, same, then a concert hall, the same, and anybody who has been around musicians playing outdoors, with no walls around, knows this. A guitar on a beach, sounds like a guitar in your living room, a trumpet in the middle of a football field sounds like a trumpet in the auditorium, and 100s of other examples. Location/venue, if it is a factor, isn't much of one. And it's the exact same thing with voice, outside or inside. In the Fall it can be checked out, in football fields all across the US, get three kids in band with 3 difference instruments that can play the exact same note in the same octave (frequency) and get the conductor to turn his back to them and play one at a time and he/she will say, trumpet, saxophone, and whatever else fits in there. flugelhorn maybe.

I think we may be talking at somewhat at cross-purposes here but also have one point where we disagree, on the importance of localization. There is no doubt we can tell instruments apart from each other. In the real world, part of what lets us distinguish them is always different physical locations (including the locations of our ears!), apart from instances where obfuscation occurs, and I can only think of two: rooms and speakers. In concert halls (but not only in concert halls), our ability to localize and reliably tell apart individual instruments diminishes because of the acoustic conditions. And then with loudspeakers you can mash all of the sounds together monoaurally into a single spot, which again makes them more difficult to distinguish. Note: in both cases the key word is difficult, not impossible. (A distant third instance of obfuscation is compression by codecs. I don't know if you've ever listened to very low bit rate material. The nonlinearities are incredible. But we'll ignore this aspect.)

Briefly on masking: I assume most people on ASR are generally know that strong sounds make it harder to hear weak sounds (this is one of the reasons nonlinearities are hard to hear). Extending this a bit, music masks its own fine detail, and in concert halls or any other reverberant space (or added reverberance to the recording), you have two-way masking: reverberance masks musical content, and musical content masks reverberance. It's a perceptual morass.

Each ear individually has the ability to unmask sounds masked by other sounds because auditory info is broken up by the cochlea into many channels, which are processed higher up to reveal correlated and uncorrelated information in terms of level, spectrum and envelope (and other things I don't understand too well).

Binaural unmasking is another phenomenon which heightens the above ability because the same correlation analysis reveals differences in the sound arriving at each ear. Unless you happen to have only one functioning ear, you are always processing two primary sets of information, even listening to one instrument.

Physical location has another important aspect: directivity. Instruments do not radiate, per frequency, very evenly. Their direct sound and their reflections in that sense contain different spectral/timing information which is important for identifying them. I have anechoic recordings of an orchestra (made in an anechoic chamber). The result is... very different, "unrealistic".

Rounding all this out and getting back to loudspeakers: I think the reason they work and sound so realistic is that we are used to echoic, reverberant conditions which complicate localization and ambiguate (would you look at this new verb I made) timbre.

Do loudspeaker nonlinearities have any benefits for timbre? Distortion all sounds kind of same-y to my ear, but then this is the first time I'm asking myself the question with all of the above in mind.

 

[Travis]

think we may be talking at somewhat at cross-purposes here but also have one point where we disagree, on the importance of localization. There is no doubt we can tell instruments apart from each other. In the real world, part of what lets us distinguish them is always different physical locations (including the locations of our ears!)

Well it’s easy to seem like it’s at cross-purposes, because I tend to narrow in on one issue and have tunnel vision to the exclusion of other factors, and the big picture, Certainly don’t intend to be at cross purposes, as my purpose is the learn, return that where I can, and correct misinformation on what is generally known.

Where I started on timbre perception was answering someone’s question about the ability to distinguish instruments.

Localization doesn’t play much in timbre perception, that I have read, but I do know that it’s a major factor in trying to obtain the most accuracy in audio reproduction through loudspeakers. No doubt it it’s key. I haven’t even dug into that specifically, but do know that the Psychoacoustics Technical Section of AES, chaired by Dr. Olive has a high priority on research associated with spatial audio.

Probably 90% of what we know about how to accurately reproduce sound with loudspeakers, came from the development of the telephone, Bell Labs. I will post the papers from the Symposiums about 1932ish) in a new thread in the technical library section and link it here. They established, to have the most accurate reproduction on localization, that what we would come to call stereo should have three speakers, LCR. Three was nearly as good as 4 or 5 and three was the number considered good enough. From there the industry, well we all know we ended up with 2 channels for stereo.

 

[kemmler3D]

Really late to this thread, but answering the original prompt:

Yes, definitely.

In music production, nearly every instrument gets at least a little "warmth" added from the original dry recording or synthesis - in technical terms, harmonic distortion. It's considered euphonic, @Rja4000 's comparison to salt and pepper is very apt. Almost everything is better with a touch of distortion (or salt).

However, all or >99% of the distortion should have been added in the studio. Just like in any decent restaurant, the chef will have already added the proper seasoning before sending you the dish.

Philosophically speaking, I think it's inappropriate to try and add more later on, just like it's considered boorish to add salt to the food in a nice restaurant.

That said, you may like the sound of more HD despite my personal discomfort with the idea.

 

[kemmler3D]

Do loudspeaker nonlinearities have any benefits for timbre?

It's possible, but would vary case-by-case.

I would say physical location might matter for timbre in extreme cases. But orchestral instruments have been developed over literally hundreds of years to have a set of distinct timbres that retain their unique character in wildly different situations, so it's (intentionally) hard for us to confuse even a saxophone and a clarinet.

There's nothing special about the concept of timbre that makes it hard for the ear to fail in perceiving it, above any other type of harmonic content.

It's worth noting that the cues the ear uses to determine location (ITD, ILD) don't have much to do with harmonic content or masking AFAIK, so I would agree that one shouldn't influence the other very much in normal circumstances.

 

[MattHooper]

This is something I've pointed out before...

When the subject is whether anyone would *prefer* audible distortion added to their playback, for instance a tube amp with audible distortion, it's very common from the accuracy-first folks to say "Well, ok, one might like some distortion for some content, but if you add the same distortion then only some stuff may sound good, other tracks will sound worse."

Basically: The effect will be variable with recordings: some will sound good, others won't.

But that is precisely the situation someone is in with an accurate system!

Due to the variation in sound quality in recordings, an accurate system is going to expose bad and good recordings, so your experience of "good sound" will vary across recordings. Some will sound worse than others, some better.

So there is no "win" in that regard for the accurate system in terms of the argument above regarding a preference or not for some distortion.

As it happens, like I stated before, I tend to like the tube distortion across pretty much all content. I prefer to have as many recordings sound good to me as possible, and I find my tube amplification seems to work very well in meeting this goal. (Similarly, I'm sure that some people who've gone for strict accuracy can also enjoy music across a wide variety of recordings. Ultimately that comes down to the individual).