What about ADSR envelope of the tones transmitted by pickups?

[bbsailor]

Yes, the high frequencies are damped quickly, and so they are important for the attack and decay. In fact "ADSR" is kind of artificial: the A and D go together as the transient. I think the reason that they are separated out is that one way to synthesize a signal is to construct a set of fundamental and overtone ratios and then complete the sound by modulating them. You need to specify both attack and decay since this is something you are imposing on the signal, not something you are just letting happen. (About 35 years ago I thought a different way to make a synthesizer would be to solve the differential equation describing an instrument, be it real (and thus approximating reality at some level) or artificial (and capable of "unreal" sounds). Then the ADS would just happen, and you would have a related equation for the R. Computers are probably just about up to doing this now.)

Mike, Mark, Steve

I believe we are talking about three things here.

1. The human ear is very sensitive to the initial transient of the string response as it gets translated by the (2) pickup and then the (3) amplifier/speaker.
2. The pickup must translate the initial string transient where the motion is the greatest and the full range of the string mechanical motion is captured and not attenuated by either the magnetic field limiting the most extreme string movement or pickup-related seen capacitance that reduces the slew rate. Slew rate - Wikipedia, the free encyclopedia
3. The amplifier and speaker must be able to fully capture the full range of the initial string transient and it's harmonics and reproduce it through the speaker.
See an excellent explaination by Ken Gilbert. slew rate explained | ken-gilbert.com See his explaination of capacitance on how it limits slew rate in general.

Here are some points to ponder. If one were to model a perfect string transient, what initial voltage would that peak string motion represent to capture the full movement of the initial attack? How do different pickup designs (magnetic field, inductance, capacitance) affect the transduction of the mechanical motion of the string? How much amplifier headroom is needed to fully capture the initial transient and not affect the subtle audible aspects of the initial transient. The answers to these questions will begin to answer the relationship of these variables to what we hear.

Joseph Rogowski

[Mark Hammer]

Here's a quarter twist...

Of late, I got interested in the applications of a conceptual approach to human audition established by a former prof of mine; the prof being Al Bregman, and the approach being what he termed Auditory Scene Analysis. You can find a synopsis of the approach here: Al Bregman's Website

The concept of ASA approaches hearing much as a Gestalt psychologist approaches vision: how do we take the noise landing on two eyes (or two ears in the case of hearing) and parse/group it into a world of objects and energy sources? What principles do we automatically apply to do that, and what portions of itare essentially "givens" of the nervous system wiring? We only have two ears, and in most instances, any given sound source will reach both of them. How does all of that get turned into THIS sound coming from over there, those two over there, and this source starting before that one?

I raise this because the whole issue of note envelopes, as detected by a pickup, is also a psycho-acoustic challenge with respect to being able to detect multiple note envelopes from the same source (as well as being abkle to tell that source apart from all others). We rarely energize all 6, or 12 (or 4 or 5 or however many the instrument has) strings in perfect mechanical synchrony. They are energized by a pick or strum, or even a slap, slightly staggered. The cognitive/perceptual task is to be able to identify each string as a separate sound source, based on the detectable properties of the spectral content they produce (spectral distribution, amplitude, timing, etc.), and effectively assign it a detected envelope.

As with any aspect of hearing, the neuropsychological properties of the hearer meets the physical properties of the stimulus to be heard halfway. Stated another way, one needs to take into account what the instrument itself physically provides that can support the perception of individual note envelopes, given how the human hearing apparatus works.

I'm biased because he was one of my favourite profs, but I think this whole area is worth looking into by pickup makers. For example, when we read reviews of pickups or guitars, reviewers will often wax poetic about "good note separation". What IS that, what makes it possible, and how do we - as hearers - achieve it?

[dor baruch]

thanks Mark Hammer
I'm going to do research on "good note separation"
i have Pretty good Understanding on psycho-acoustic and spatial perception of sound (as a sound and acoustic engineer)
Just remember that low-frequency sound waves behave and spread in a spherical way in space and as higher the frequency the wave is spread in a more linear and straight way
i know that when the pickup read alot of low freq and not enough hige freq or overtone and you combine several simultaneous sound vibrations string the resulting sound is as pepole say "bad note separation"
i think that the hige freq Compared to the rest freq given Significantly contributes to our perception of sound separation

[Mark Hammer]

As Bregman notes, it is the many elements that we can package together that lets us perceive a sound as coming from a different/separate source or object. Whenever we have the annual debate about capacitor types here, I hasten to remind people that cap properties start to become more important when we are listening to mixed-down multi-source acoustic music, since we are faced with the challenge of lining up, "packaging", and assigning all that harmonic content to the appropriate source. Anything that interferes with the ability to assign THIS harmonic to THAT source, via group delay, makes the music sound more smudged and less realistic.

So what do the combined properties of a pickup, string, and bridge system bring together that can let use hear 6 separate note sources with ease. I emphasize "with ease" because perceptual note separation can always be achieved when the circumstances are not difficult and cluttered. What leads reviewers to remark is how easily notes are perceptually separated, even under "busy" conditions.

A digression: Some 30 years back, I bought a couple of the "stereo image enhancers" that had become popular amongst audiophiles at the time. These were essentially bucket-brigade-chip based devices that would artificially create "sound shadows". The principle was exotic then, but simple now. Subtract right from left channel, and left from right channel (keeping in mind there was less channel separation on vinyl than there is on CD or MP3), and feed each of those derived signals through a brief delay of a few milliseconds. Now impose a little bit of lowpass filtering, and crossfeed the delayed derived signals to the opposite channel. In this way, content that was unique to the left side (remember, we've subtracted anythng shared with the right channel) would come through on the right channel, but a little bit later and duller than when it appeared on the left channel.

This was a method to go beyond what simple adjustment of left-right amplitude balance did, with respect to situating something in the mix as being "over here" or "over there". We've come a lot farther, using digital delay algorithms in Q-Sound and the SRS that Windows users are familiar with, but even the early analog approaches provided the sorts of audio cues that made listening MUCH easier, because spatially locating something in the mix when you closed your eyes (the congas are riiiiigggght....THERE) was easier. It increased hiss and crackle (because noise is unique to each channel and the devices increased the overall amplitude of whatever was unique to a channel), but it still made it easy to listen for long periods of time because the mental work required to do the "auditory scene analysis" was reduced. Like the difference between trying to read big crisp typeface on crappy newsprint vs trying to read a photocopy of a photocopy of a photocopy, made by someone with Parkinsonism, on top quality paper; it's the ease with which the typeface can be read, not the degree of background noise, that matters (within reason).

So, by analogy, what is it about the mechanics and electronics of guitars and pickups that lets us mentally "hear" 6 separate notes/strings with the least amount of mental effort, even when that soundfield is cluttered by repeated unsynchronized transient attacks, and even distorted electronically?

I think there is value for us in adopting a cognitive neuroscience approach to understanding electronically processed sound.

[Steve Conner]

For me, the fact that people claim to hear differences between guitar tone capacitors in Youtube videos made with camcorder mics is proof that you are imagining it all.

[Mark Hammer]

Well, I didn't want to start another capacitor war. My point was to say that the starting point is to ask what the human mind has to accomplish in any given auditory setting, what might get in the way of that, and what might assist it. For my part, I think the rock and roll guitar context is largely irrelevant to discourse about caps, but situations like I described (reproduction of multi-source acoustic music) can be impacted on by capacitor qualities.

But yeah, strong claims about ANYTHING demonstrated with a webcam mic on Youtube is unlikely to be the basis of cogent reasoned debate!

[Steve Conner]

Well, my personal feeling is that the human ear is a pretty low-resolution instrument, but the brain does a wonderful job of constructing colourful illusions to fill in the gaps, feeding off all the available information. The illusions can be destroyed by blind testing, but you're left with a deathly dull experience that no creative type would put up with. So nobody does it, and 99% of the banter on the internet is arguing over the difference between different peoples' illusions, not over any actual physical phenomenon. I refuse to participate any more (except for making smartass posts like this )

[Mike Sulzer]

Well, my personal feeling is that the human ear is a pretty low-resolution instrument, but the brain does a wonderful job of constructing colourful illusions to fill in the gaps, feeding off all the available information. The illusions can be destroyed by blind testing, but you're left with a deathly dull experience that no creative type would put up with. So nobody does it, and 99% of the banter on the internet is arguing over the difference between different peoples' illusions, not over any actual physical phenomenon. I refuse to participate any more (except for making smartass posts like this )

Actually, I think what Mark described above makes a lot of sense. The human ear/brain system is very sensitive and flexible, but like a lot of things, it does not work in the most obvious way. Sampling a set of filters in the time domain has its limitations, and it is remarkably complicated because the relatively slow operation of nerves means that the analysis is different at low, medium and high frequencies.


As for capacitors, bunk is bunk, always will be, and always will be with us. There are plenty of interesting real things to investigate.

[jason lollar]

Most of us are guitar players with no synth experience but I own several vintage synths I gig with. Anyway its amazing what you can do with ADSR. Guitar pickups on the other hand you have alot of different ways to build a pickup that effects ADSR - I never thought about comparing a pickup to the envelope controls but its an interesting concept that I have subconsiously been using for years. not many people talk about how a pickup feels- I have been aware of that for decades but its also dependant on the amp and the guitar. However-a pickup with a steel core is different than a pickup with an alnico core and different grades of alnico have different attack properties partly due to inductance differences and partly due to magnetic strength- my experience talking not any kind of formal training but its pretty obvious and easily tested but not many people take the effort to remove all the variables and test one thing at a time, this is where having an auto feed winder of any kind- mechanical or computerized can really help you compare A and B with consistancy.
I wont hand out information you can easily find out for yourself but its an interseting observation- I dont know how R fits in with a guitar but ARS certainly do. Thanks for bringint it up. I have used those terms before but never saw it so clearly DOH!

[jason lollar]

Most of us are guitar players with no synth experience. Anyway its amazing what you can do with ADSR. Guitar pickups on the other hand you have alot of different ways to build a pickup that effects ADSR or at least ADS - I never thought about comparing a pickup to the envelope controls but its an interesting concept that I have subconsiously been using for years. not many people talk about how a pickup feels- I have been aware of that for decades but its also dependant on the amp and the guitar. However-a pickup with a steel core is different than a pickup with an alnico core and different grades of alnico have different attack properties partly due to inductance differences and partly due to magnetic strength- my experience (not any kind of formal training) its pretty obvious and easily tested but not many people take the effort to remove all the variables and test one thing at a time, this is where having an auto feed winder of any kind- mechanical or computerized can really help you compare A and B with consistancy if you have control over every other aspect and know what the variables are.
I dont know how R fits in with a guitar but ADS certainly does. Thanks for bringing it up. I have used those perameters before in a descriptive way but never saw it so clearly DOH! with a synth you can see it on a scope, I imagine a guitar would be harder to see a wave form like you can on a synth- I will have to try it

[Mark Hammer]

And keep in mind that the string/pickup interface is a bit like the old Fairlight synthesizers, where you could use a light pen to draw the envelope of each successive harmonic, such that harmonic X rose and fell over the lifespan of the note THIS way, and harmonic Y and Z rose and fell THAT way. That is, the envelope of the note is a multi-component event, not a uniform rise, fallback, hang-in-there, and die out. It is the activity at the start of a picked note that easily differentiates a Tele bridge pickup from a P90 on a LP Jr. The physical aspects of the guitar design also play a big role i providing what there is to detect by the pickup, but the pickups also play a role in making the first 100msec of THIS pickup/guitar sound different from THAT one.

[David Schwab]

The physical aspects of the guitar design also play a big role i providing what there is to detect by the pickup, but the pickups also play a role in making the first 100msec of THIS pickup/guitar sound different from THAT one.

I think you will find that the guitar makes more of a difference. What's the difference between a Tele pickup and a P-90? They are both single coils. They have different shaped coils and magnetic circuit, but sticking a Tele pickup on a LP Jr wont make it wound like a Tele, and vice versa. It just filters the tone. The main cause of this is the scale length of the guitars. We have also all heard humbuckers on Strats, and they still sound like Strats, only thicker. Longer scales are twangier, and shorter scales have a rounder tone.

IMO the only difference a pickup will have on attack is where the resonant peak is. An upper midrange peak sounds sharper, and seems like it has more attack, because it's louder where our hearing is most sensitive. An electric bass can have a stronger attack, and more sustain too, because the strings are longer. But it wont sound like a Tele because it wont have that sharp treble bite. And of course the strings make a big difference.

But the pickup? Not really. That would be like having different microphones drastically changing the singer's voice. They don't, but they will EQ it different ways.

[David Schwab]

The physical aspects of the guitar design also play a big role i providing what there is to detect by the pickup, but the pickups also play a role in making the first 100msec of THIS pickup/guitar sound different from THAT one.

I think you will find that the guitar makes more of a difference. What's the difference between a Tele pickup and a P-90? They are both single coils. They have different shaped coils and magnetic circuit, but sticking a Tele pickup on a LP Jr wont make it wound like a Tele, and vice versa. It just filters the tone. The main cause of this is the scale length of the guitars. We have also all heard humbuckers on Strats, and they still sound like Strats, only thicker. Longer scales are twangier, and shorter scales have a rounder tone.

IMO the only difference a pickup will have on attack is where the resonant peak is. An upper midrange peak sounds sharper, and seems like it has more attack, because it's louder where our hearing is most sensitive. An electric bass can have a stronger attack, and more sustain too, because the strings are longer. But it wont sound like a Tele because it wont have that sharp treble bite. And of course the strings make a big difference.

But the pickup? Not really. That would be like having different microphones drastically changing the singer's voice. They don't, but they will EQ it different ways.

[dor baruch]

That would be like having different microphones drastically changing the singer's voice. They don't, but they will EQ it different ways.

I think its the best way too Interpret this subject

[Steve Conner]

Yes, but bear in mind my point about the two orthogonal sets of vibration modes. The singer would have to have two mouths, if only one set of vocal cords.

You can imagine that a double-headed mic that picked up both mouths would sound different to a mic that sensed only one or the other, and the difference wouldn't just be a matter of EQ.

Say one mouth sang "Ooooo" while the other sang "Yeah". Ooooo has the same formant throughout, but in Yeah it changes three times. No amount of EQ will transform one into the other.