Fragile Harmonics - Not so crazy after all? =D

[Max]

Hi,

The question which kind of guitar tone, timbre, sound, or how ever you want to call it, is perceived as being "better" - be it ("better") in the sense of :

- just more pleasing all by itself, like many will perhaps prefer to listen to a nightingale than to a crow
- in the sense of "a more appropriate instrumentation of a certain song"
- in the sense of "more up to date and cool"
- etc.

could IMO be answered today in a rather precise scientific way, just like the question can be answered today with a rather high precision, if certain groups of people prefer the taste of Budweiser or Jever.

First class psychoacoustic laboratories like the one of Bose http://www.bose.com/controller?event=VI ... /index.jsp or Fraunhofer http://www.iis.fraunhofer.de/bf/amm/dow ... 209_EN.pdf

today can answer questions concerning which kind of sound people perceive as being more desirable for the engine of their car or the sound of their TV etc. with high precision.

And a first class psychoacoustic laboratory is not only able to research what people perceive as being "better", but to find out exactly what it is precisely what they like better, what measurable physical data match descriptive terms like "sharp", "warm", "piercing", or "straining". And they can even find out how large the gap is between the perception of two guitar sounds and if this gap in perception is of a different size if people know that they listen to one sound recorded with a Dumble and an other one recorded with a Dumble clone, or if they don’t know this. And they can find out, as an other example, if there is a significant difference between the perceptions of guitar players and those of usual listeners in an audience.

IMO the question "What kind of sound emitting device is perceived as being the better one?" cannot be answered just by researching the measurable physical characteristics of different sound emitting technologies and without taking into account all the biological, neurological, psychological, cultural, etc. factors that steer the human sound perception process.

So IMO we shoudn’t confuse - just as an example - what the terms "soft clipping" and "hard clipping" mean in a measurable physical sense and context with the kind of descriptive terms that people use to describe their perception of some sound as being "soft" and "sweet" or as "hard" and "scratchy". IMO metaphors sometimes can be rather misleading in scientific research.

Of course everyone who develops technical devices that address the human perception, be it for entertainment or information or whatever purpose, IMO should at best know how human perception works. An engineer who wants to develop "the best" 3D home cinema system of course has to know how and why and when precisely steroscoping viewing leads to the subjective perception of "depth".

That is why in psychoacoutics many different sciences, like physics, biology, psychology, neurology, test theory, statistics, computer sciences, etc. work together to help answering questions like "What codec should be the successor of mp3 to improve the fun when people are listening to digitalized music and the ease of storing and broadcasting it in a most appropriate way?". And - at least in a general sense - this is IMO not another kind of question than "What kind of guitar amp should we build to improve the fun of whatever special kind of player or audience?" or perhaps "How can we offer the same amount of fun to musicians who only can pay $300 for a guitar amp?"

But I admit: Psychoacoustic research isn’t inexpensive and may only make sense and be available for larger companies. But on the other hand perhaps the psychoacoustic laboratories of some universities may even be interested if a "boutique builder" would ask them to set up some research project together. And at least some books or lectures about psychoacoustics and its applications concerning the development and validation of musical instruments AFAIK wouldn't cost a fortune.

@talbany,

Tony, IMO you are right: Even if you would have a transistor amp that would sound exactly like #102, it would IMO be nearly impossible - at least at this point in time - to sell it for the price of an Ojai as an example. If the transistor #102 would cost a tenth of the tube amp – then perhaps – only as a "practicing amp" of course.

The only way to get perhaps enough credibility to be able to sell it for the price of an Ojai would IMO be to convince RF to replace #102 by this transistor amp.

Here's some additional info concerning the Fraunhofer HD-AAC codec for those here interested in such a topic:
http://www.youtube.com/watch?v=JESMLDu_AyM
http://www.iqmagazineonline.com/article ... cle_id=697

Cheers and all the best,

Max

[talbany]

A pretty good example of FASHION!!
One of the most popular SS amp simulators of my time has to be the Rockman.. I remember when these hit the market as they had a huge impact on all types of players (that didn't seem to mind that it didn't sound and feel EXACTLY like a tube amp) and quite possibly IMHO had the momenteum and backing to change how we looked at tube amps.. (BTW didn't Todd Sharp use one in the studio w/Christie Mcvie after recieving 2 custom made Dumbles and matching cabs)..I remember just about everyone had one or wanted one..Tom Schultz even had some of the most popular players talking about and even using it..Here is a pretty impressive list for the time..

Tom Scholz of Boston
Steve Stevens
Alex Lifeson of Rush
Charlotte Caffey of The Go-Go's
Phil Collen and Steve Clark of Def Leppard
Billy Gibbons of ZZ Top
Megadeth on the album Peace Sells (1986)
Bernie LaBarge on his CBS album Barging In (1983)
Neal Schon of Journey
Curt Kirkwood of Meat Puppets on their album Up on the Sun (1985)
Buckethead soloist mainly using rockman

Of coarse we all know the Rockman ultimately failed (although one could argue it's success)to become a permanent staple and Tom sold the company for peanuts..Why did it fail..
Perhaps not improved upon poor financial backing/mismanaged or Robben Ford decided not to use one.. Could be maybe just fell out of fashion..
I choose fashion..

Perhaps if this forum or any other were centered around the tones of Tom Scholz or others on the list above and not say a Robben Ford, LC, SRV/Dumble perhaps the perception of SS would go in a slightly different direction..Just something to think about

Disclamer.. I never owned a Rockman.. I did have his Powersoak and Chorus delay unit at one time

Tony

[Structo]

I had one of those.
If I recall I broke it about six months after I got it.

Then I bought one called a Polaris, it had distortion, chorus and reverb.
Sounded great!

The weak spots on those devices was always the PCB mounted jacks, that would break the solder joints.
You could fix them a couple times but eventually they were unfixable.

[renshen1957]

Sorry but I have to strongly disagree that digital signal processing can't accurately reproduce the subtleties of a tube amp. Anyone who is well versed in modern technologies and capabilities would realize how benign that sort statement has really become.

One of my good friends and I have spent hundreds of hours discussing and designing digital systems that can accurately emulate analog without sounding like, well, emulations. We call it "analog inspired digital" style design and while I honestly can't program for shit, I've spend a lot of time figuring out and studying quirks in analog and my buddy will figure out how to model whatever phenomenon in digital. While our focus is on analog synths at the moment, I can promise you that we could (and eventually will) shift our focus to modeling the strange things that happen when you push a tube amplifier into the non-linear operating regions.

The thing is that a lot of the digital stuff that has been introduced so far has been complete crap and this has tainted a lot of peoples views of what digital modeling is really capable of. It's a shame too because as hard as some try to change peoples opinions, guitar players are some stubborn bastards and it's hard to undo the damage that has been done in the past. All I can really say is give it another 10 years and I bet many people will change their tune (tube?) when it comes to digital modeling stuff.

Hi,

Although not as clearly written as I would like to have (I posted past midnight my time after a long day's work and sleep deprivation from the night before), I think I will stand pat on my statement that Digital Simulations to date have promised much and delivered less than the hype.

I agree that with another decade there will be advances in digital modeling.

Whether these advances will be up to the task to change tube amp adherents is another matter. After all if it isn't broken, why fix it? When one's tone signature is perceived as what makes one unique or what is keeping you employed and paying the mortgage, I can understand the tenacity of Guitar players and tube amps.

You wrote, "I have to strongly disagree that digital signal processing can't accurately reproduce the subtleties of a tube amp."

Well, the track record for digital signal processing hasn't been that successful to date. At best it would be a D- for tube amps. Maybe C for digital instruments as I will comment below.

You wrote, "While our focus is on analog synths at the moment."

As someone who worked with (an occasional on) analog synthesizers in the 1970 (Moog IIIP, Mini Moog, Buchla etc), these devices were all Solid State. (No company to my mind produced a classic analog tube synthesizer for retail sale used by bands that I am aware off.)

I have admiration for your accomplishments, emulating a solid state device with another solid state isn't exactly the same thing as emulating either tubes or for that matter acoustic instruments.

The general impression in related area of digital reproduction and/or the use of digital sampling is that no one to date has built a digital piano that among blind listening tests (to eliminate the psycho-acoustic factor) couldn't be distinguished from the real thing amongst pianists. I could say they same about sampled sounds (software for digital organs for $100,000 installations in Churches), and Harpsichord samples off 18th Century instruments. I've played Pipe Organs and digital organs) and as audience member (organists are diehard friends and will fill the chairs at each others performances) as well as the latest digital harpsichords software and real harpsichords, and my conclusion is that pipe organ builders and harpsichord makers aren't going to go extinct in the near future (unless the economy keeps being depressed) or in the next decade for that matter.

(I would extend the above observations to the best software for orchestral instruments. At least I can tell when a score has real instruments or canned ones.)

As to the quality of the sound (tone replication) among the players of digital pianos and organs, the consensus is these instruments are used for their expediency and convenience not because tonal signatures had been cloned to a tee. Ask any major keyboard player and they will say it beats trying to transport or rent have delivered a grand piano, a synth, and an organ to a gig. The equivalent of a Line 6 amp for cover bands.

Ask if it is indistinguishable from the real thing, and the replies are essentially a firm negative. For example, when I last talked to Goldy McJohn (yes, the Goldy McJohn) about his keyboard he was playing (it produced a good replica of a Hammond B-3 in instrument about the size of a computer keyboard), he replied that it got the job done was good because he couldn't get the roadies to deal with transporting and lifting a real Hammond (tube model). He said he would gladly switch back to the real deal in a heartbeat (he wasn't 100% convinced of the digital replication) if could find a practical solution to transportation. To his credit, he made the keyboard sound very close to the original.

Electro Harmonic's Holy Grail emulation of the classic tube Spring reverb by Fender is advertised as "so faithful that even Dick Dale couldn't tell the difference." (The Grail is nice, for sure. But after 40 years of loud rock music and chemotherapy for cancer, I'm not sure Dick Dale's hearing was in such great shape when the Grail came out.)

At least among the guitarist I know (my son included) digital delays of any type are used as a separate effect on its own for its flexibility. Many eschew using the "analog" settings and use the real deal as a separate effect and especially on recordings. And those Tel Ray Oil Can delays have their proponents, too, although I do not know of any digital simulations.

Ask a Concert Classical Pianist about digital and they will flat out say digital does not "cut it" on terms of performance being exactly equal and dynamic as an acoustic piano. Even with the instruments with hammer touch keyboards, the tone wasn't indistinguishable from the real thing or produced the overtones (even without the complication of the damper pedal being depressed) in the same manner as a piano.

An excerpt from review of the state of the art Yamaha sums this up:

"key-notes Home > The Piano > Yamaha Digital Piano
Yamaha Digital Piano
A Pianist's Review of the AvantGrand

I recently had an opportunity to play the AvantGrand, the new Yamaha digital piano that represents the current state of the art, and wanted to share my experiences with it with key-notes readers.

First, as a classical pianist I must caution that no matter how good digital piano technology gets, nothing replaces the real thing, and I don't see that changing even in the distant future. After all, Stradivarius and Guarneri violins have only become more valuable over time, and the best modern technology has been unable to match the beauty of their sound. Where digital versus acoustic pianos are concerned, a recording of a note (called a "sample") triggered by pressing a key and played by an electronic speaker just isn't the same as a real hammer striking a real string causing a real soundboard to vibrate."

Maybe Classical Pianist are stubborn bastards, too (my wife thinks so of me, but I was stubborn before I took up the Piano, Harpsichord, and Pipe Organ) I don't know of any name Classical Pianist that would perform a recital on a digital piano or have given a product endorsement to date for one (and Yamaha has lots of money. Maybe Classical Pianists aren't endorsement whores, too?)

You also wrote, "I can promise you that we could (and eventually will) shift our focus to modeling the strange things that happen when you push a tube amplifier into the non-linear operating regions."

Tubes aren't all that linear a device to begin with in their "linear" operating regions, and when pushed into distortion and other non-linear regions are even more complicated to understand and chart these variables.

Although the physics of SS components are very well understood, where tubes are concerned, we have a general concept and idea how they work, but being electromechanical devices, no one has fully worked out a complete understanding of why there can be as much variance as demonstrated in a tubes performance, even among tubes of the same dates of manufacture of the same type (12AX7). (I won't go into component chemical variables, as the late Ken Fisher commented on in a interview)

As I have said previously, (though not in yesterday's post):

How does a tube react to AC signals and at what frequency, and at what voltage, and at what current levels? (Highly variable.)

Add in transconductance, Mu, internal plate resistances, Miller capacitances, plate currents, as variables, and how tube behaves (or misbehaves) when overdriven (THD and IMD).

Throw in the minimum voltage below which a tube cannot pull its anode toward the plate and what part of the tubes transfer curve is being utilized at low voltage signals as opposed to high voltage signals (even when controlled by a negative feedback loop) and phase changes from incomplete signal inversions.

Also toss how a tube reacts to supply voltages which are not rigid (voltage sag) in the power transformer, interaction between the output transformer (and the power tubes): reflected impedance and other factors such as leakage inductance and vice versa), at what point and when an output transformer saturation occur. (I won't even start on the quality or lack there of concerning the Iron used in the laminations. As to the argument over the insulation, Paper or Plastic, anyone?)

These are just a few of the parameters that would have to be quantified, and I am only scratching the surface.

I haven't addressed components as variables, such as capacitors (resistance, as well as inductance, and distortion present in Electrolytic caps), and shan't go to deeply in detail for this evening do to constraints on my time. To briefly touch on the subject of capacitors, (Dumble builders use ceramic caps) ceramic caps exhibit variation of a capacitor’s impedance with frequency variation. As to a Tweed Bassman model, the carbon composition resistors' resistance on the plates actually varies with the voltage across the resistor.

And if one is trying to model a specific type of amp, then toss in the speaker cone's behavious, sound pressure levels, open back cabinet/closed back cabinet, room acoustics (standing waves), whether the speaker has an issue of beaming. These vary with signal intensity, signal volume, room temperature, humidity and barometric pressure.

Again, I have only addressed the tip of the iceberg.

If after doing so, one would have to use a Hi-Fi amp with sufficient headroom (lots of Watts) and a neutral full frequency speaker with minimum coloration to the final signal without adding any additional artifacts to the digital ones produced (hopefully at an inaudible level).

I sure wouldn't want to model a Dumble ODS or even Marshall 800.

The architect Ludwig Mies van der Rohe who said "God is in the details," also said "A chair is a very difficult object. A skyscraper is almost easier. That is why Chippendale is famous." I wonder what he would have said about linear devices (SS/digital) attempting to replicate a non-linear device (tubes amps)?

Maybe in ten years time the Super Computers of today can be made faster and shrunk in size to eliminate latency (using nano-vacuum tube technology?) to facilitate an economical and super minuscule size to crunch the numbers of these and a host of other variables to facilitate replicating a tube amp and be faster than Damn Fast. (And maybe include random numbers generator to facilitate the randomness of tube behavior?) The price would have to be affordable as 90/10 rule of money applies to Guitarists (as well as keyboard players) too.

Maybe if all these factors can be digitized, the proof will be in a touch responsive amp that can mimic a tube amp.

Hearing will be believing.

Then again, as my wife points out when see asked why I am writing this email, "Wouldn't just be easier to use a Tube amp? I use ginger in recipe that calls for ginger."

Best of Luck in your future endeavor.

Steve

[Max]

blind listening tests (to eliminate the psycho-acoustic factor)

Hi all,

Because "the psycho-acoustic factor" of course never can be "eliminated", because this would "eliminate" hearing something at all, IMO this statement showcases that even knowledgeable and experienced members here obviously have a complete misconception concerning what is meant by "psychoacoustics". IMO this is a bit weird, taking into account that what the Amp Garage forum is all about is building musical instruments.

So for members here who might be interested to get at least some idea what really happens in the every day research reality of "psychoacoustics", here are three papers:

The 1st paper is entitled: "Music and Psychoacoustics":

http://psychology.jrank.org/pages/1528/ ... stics.html

The 2nd paper is a (IMO) very interesting NASA paper concerning not a "surround" but a true "3-D Sound for Virtual Rality and Mulimedia":

http://www.macs.hw.ac.uk/modules/F24VS2 ... imedia.pdf

If you should only be interested in getting just a first impression and overview concerning "psychoacoustics", I would recommend reading at least the "Table of Contents" (pages v – viii), the part entitled "Perception" (pages 25 – 29), and the part entitled "The Music of the Future is the Music of the Past" (pages 193 – 198) of this NASA paper.

BTW: This NASA paper may perhaps be an interesting read of even some general interest, taking into account that listening to music by head- and earphones is rather common today anyway.

The 3rd paper is entitled "Timbre Models of Musical Sounds":

http://www.aaue.dk/~krist/TMoMS.pdf

Cheers,

Max

[rogb]

What a great thread!

I recently dug up some Guitarist and Guitar Player mags from the 90s and it's all about modelling and simulation back then.

Still got my Sansamp GT2.. can't bring myself to sell it

[Luddy]

I teach at a University with good psychology and good speech and hearing programs. My students have assisted in music cognition tests, so I know that some work is being done in music. We also have an anechoic chamber, which I have used to host 22 instrument recordings online. The link is http://theremin.music.uiowa.edu/MIS.html

If anyone has something that would be an impact and that would benefit from our resources, let me know.

[Cliff Schecht]

I've done my fair share of research WRT the human ear, both personally and in various projects I've worked or am working on. I've got access to a couple of really nice acoustic chambers as well as lots of instrumentation (some of which I've designed) that is used to measure how the brain responds to certain stimuli and how this can be used to diagnose various diseases (ranging from schizophrenia to vertigo). It's always fun to do this type of research too. Learning so much about the human ear has had an influence on how I do any audio-based design work and even how I play music in general, especially when it comes to requisite frequency response, minimum needed gain steps and especially maximum tolerable SPL (which is why I never build anything over 50W ).

[dreric]

as well as lots of instrumentation (some of which I've designed) that is used to measure how the brain responds to certain stimuli and how this can be used to diagnose various diseases (ranging from schizophrenia to vertigo).

Hey Cliff

This is an area of interest for me. If you don't mind, can you elaborate on how this instrumentation measures brain response. (functional MRI w/ isotopes? Active PET?) Pre-synaptic, synaptic or post- synaptic / functional? What stimuli (sound, touch, olfactory?) What are you measuring and how?

How can this being used to diagnose? Ultimately, diagnosis names for conditions are typically Latin or Greek derived description of the doctors observation of the patient and the a patients subjective description of their state. Is this instrumentation measuring a neurological correlative within brains of those suffering? How can this tech be used to diagnose?

published?

On another note I'd like to point out two concepts that I find painfully absent whenever the discussions of psychoacoustics or why certain amps are purported to sound better than others.

Is essence psycoacountic research boils down to stimulating a subject with sound and measuring their response objectively (salivary pH, respiration, heart rate, BP very much like a polygraph) and their perception of the sound. Then their response is correlated to creating a product that consumers will hopefully buy.

One missing piece in discussing this is not so much what the brain does when subjected to a stimuli but rather what state the brain is in while subjected to the stimuli. This includes both physiological conditions such as age, genetics blood sugar levels, medications and blood oxygen levels as well as the past medical history and emotional state of the listener.

In other words, a Marshall on 10 sounds better while you belly's full, you've had a few beers and you've just gotten laid. While it sounds pretty harsh when there's no food in the fridge and she's left you for the guy who used to be your best friend.

Human research without clear definition of the subject population parameters is meaningless and most often misleading.

The other factor would would be a derivative of the Heisenberg principal which implies that your perception of the stimuli affect your response to the stimuli.

In other words, If a Rockman sounded great to you it's most likely because you bought it thinking that it would sound great. Or a more current version, certain amps sound better because we are told they sound better or we want to believe they sound better.

Just my .02

Eric

[talbany]

In other words, If a Rockman sounded great to you it's most likely because you bought it thinking that it would sound great. Or a more current version, certain amps sound better because we are told they sound better or we want to believe they sound better.

IMHO In the real world this statement maybe more true than most would care to admit..

Tony

[daydreamer]

I think hybrid is the way to go if convenience is the governing factor .

My Vox Tonelab LE uses the 2 sides of a single 12ax7 to replicate the preamp and power amps of various amp models. It can sound like a tube amp because in some limited sense is a tube amp. BUT, it is limited to 24 bit 48khz processing, which is theoretically enough to fool the ear, though overall it doesn't, and on top of that it doesn't interact with the speaker like a real power amp does.

Recorded it sounds ok, but has that slightly 'processed' sound, because, well, thats what it is, a processor!

The Tonelab is an interesting unit as it has been around since the 90's when it was developed by Korg. Since then it has been in constant evolution in various forms.

This all reminds me of an article I read about japanese doll makers who make 'anatomically correct' fake women for clients. Sure the latex and silicon has a similar feel to real skin, sure it bends at the knees etc, but it is still just a doll for men who aren't getting laid for whatever reason...I'm not sure what the moral of the story is, maybe something along the lines of 'Don't give up on the real thing'!!

No doubt sometime in the next 50 years advances in robotics and doll making will make having a real human partner optional, then those fake humans will decide that we are obsolete and too much work to maintain.

I get the feeling sometimes when I'm messing around with sound equipment that it's all just an excuse not to be writing music, because well, that would be hard work. The only people who end up benefiting are those selling me the latest gadget.

I've come back to tubes for my main sound, because everything digital seems to create more things to think about, but less BEING creative.

Sometime subtle differences make all the difference.



Andy

[Cliff Schecht]

as well as lots of instrumentation (some of which I've designed) that is used to measure how the brain responds to certain stimuli and how this can be used to diagnose various diseases (ranging from schizophrenia to vertigo).

Hey Cliff

This is an area of interest for me. If you don't mind, can you elaborate on how this instrumentation measures brain response. (functional MRI w/ isotopes? Active PET?) Pre-synaptic, synaptic or post- synaptic / functional? What stimuli (sound, touch, olfactory?) What are you measuring and how?

How can this being used to diagnose? Ultimately, diagnosis names for conditions are typically Latin or Greek derived description of the doctors observation of the patient and the a patients subjective description of their state. Is this instrumentation measuring a neurological correlative within brains of those suffering? How can this tech be used to diagnose?

published?

On another note I'd like to point out two concepts that I find painfully absent whenever the discussions of psychoacoustics or why certain amps are purported to sound better than others.

Is essence psycoacountic research boils down to stimulating a subject with sound and measuring their response objectively (salivary pH, respiration, heart rate, BP very much like a polygraph) and their perception of the sound. Then their response is correlated to creating a product that consumers will hopefully buy.

One missing piece in discussing this is not so much what the brain does when subjected to a stimuli but rather what state the brain is in while subjected to the stimuli. This includes both physiological conditions such as age, genetics blood sugar levels, medications and blood oxygen levels as well as the past medical history and emotional state of the listener.

In other words, a Marshall on 10 sounds better while you belly's full, you've had a few beers and you've just gotten laid. While it sounds pretty harsh when there's no food in the fridge and she's left you for the guy who used to be your best friend.

Human research without clear definition of the subject population parameters is meaningless and most often misleading.

The other factor would would be a derivative of the Heisenberg principal which implies that your perception of the stimuli affect your response to the stimuli.

In other words, If a Rockman sounded great to you it's most likely because you bought it thinking that it would sound great. Or a more current version, certain amps sound better because we are told they sound better or we want to believe they sound better.

Just my .02

Eric

Very interesting discussion and this is something we do neglect pretty much entirely. I do find myself using a lot of words to describe emotion and how things I build make people feel. It's always great to see someone open their eyes really wide when they hit the first power chord and get blasted with a lot more sound than a 6"x4" amplifier should be able to make. A smile usually follows..

The research I am doing now is on EcochG auditory evoked responses from the brain, but before we were focusing on EvestG which is sort of in the same vain. The basic idea of EcochG (and I'm oversimplifying here) is that three sensors are placed on the head (one on an ear lobe, another under the other ear and one on the forehead) and a stimulus is played into the subjects ear. The brain is supposed to react to the stimulus a certain way, essentially it gives back a somewhat predictable pattern (looks like a ~1kHz sinusoid superimposed on a 10 Hz wave). I'm actually not certain as of now as to what diseases can be correlated to different peculiarities but I know the EvestG stuff focused on how long the brain took to react to the stimulus, as well as the overall response that we get back, to determine different problems.

I can't get into the details of what I designed because it hasn't been published yet but the signals that we get back from the brain are TINY, on the order of 100nV to maybe 1uV. So far I have my design working well down to about 500nV, but I'm having issues with oscillations because the input amplifier uses a goofy current feedback topology and doesn't like super low source impedances (typical of this type of instrument amplifier). This becomes an issue because there isn't any instrumentation I know that is capable of putting out such small signals with a controllable output impedance, I had to design the damn test equipment as well as the actual amplifier. Working with gains in the 140-160dB range is fun at times but utterly frustrating at other times. I keep hitting brick walls that take a lot of discussion and head scratching to figure out. I think once we start testing in a medical environment with real-world sensors and signals we will get better results. The sensors have a 5-10k impedance which is perfect for the amplifier I'm using, unlike the goofy stuff I have to do now to even develop a usable test signal.

[Cliff Schecht]

BTW, awesome thread derailing we have going here.

[img:413:310]http://pennymaxwell.files.wordpress.com ... rail-3.jpg[/img]

[daydreamer]

LOL

Hey you can count on me for the low brow comment, if that's what you meant Cliff!? .

Your research sounds pretty intense, pardon the pun. I love studies on consciousness and the like.

Andy

[Max]

BTW, awesome thread derailing we have going here.

Cliff, I am not that sure concerning the derailing:

This "Fragile harmonics...." statement of Alexander Dumble, the topic of this thread, is of course a metaphor: As we all know harmonics aren't "fragile" and do neither die nor "survive". Now, at least AFAIK those, all the arguments that are used in discussions about why tube amps are currently preferred by most guitar players, are IMO of a similar metaphoric kind and more close to plausibility level than to scientific level.

My current impression is that there isn’t yet found a clear and proven relation between

- the measurable differences in the physical operation of tube and SS guitar amps in cooperation with usual guitar speakers
- the measurable differences between the air pressure stimuli both kind of sound generating systems produce
- the measurable differences in the perception of the air pressure stimuli produced by both kind of sound generating systems

What IMO still isn’t answered in what I would call a scientific way is:

What precisely are the most important differences in the perception of the sounds generated by tube driven and SS driven guitar amp/speaker systems?
Do these most important differences in perception relate in a statistically significant way to differences between the measurable air pressure stimuli produced by tube and SS driven systems?
What precisely are the physical differences in the operation of both systems that are responsible for the fact that they emit sound pressure stimuli that show precisely these measurable differences that result in these differences in perception?

Up to now my impression is that at least a large part of most discussions about tubes vs. SS rarely have left a Reader's Digest level and IMO showcase more prejudices than what – at least in my personal perception - seems to be scientific knowledge.

What I mean by "prejudices" is something like this (no SS example):

Around twenty years ago I was present when a professional photographer should take some pictures of two Dumble 150W guitar amps with GE 6550A tubes, one of them an ODS. Per chance a renowned guitar topics journalist writing for some kind of guitar magazine was present, too, because he needed some pictures of some other gear for this magazine. At this time Dumble amps have not yet been in fashion, so he didn’t know and recognize these amps and didn’t have any idea concerning their sound. His comment when seeing the power tubes: "What a crazy idea of an engineer to put GE 6550A tubes in a guitar amp. GE 6550A tubes sound like shit in guitar amps so these amps will for sure suck!"

IMO scientific answers for many questions like:

- what kind of perception precisely is meant by using the metaphor "bloom"?
- which precisely are the measurable physical characteristics of the sound pressure stimuli perceived and described as "bloom"?
- which precisely are the measurable physical characteristics in the operation of a certain kind of circuit/speaker system causing it to emit air pressure stimuli that are perceived as "bloom"?

still don’t exist for whatever reasons.

Current argumentation in regard to questions like this is IMO still far more of some kind of plausible rationalisation for having spent what female fashion victims will probably perceive as way too much money for toys-for-the-boys that would far more urgently have been needed for Jimmy Choos.

BTW:

Did someone here notice the short note ("New Gear") in the December 2010 issue of Guitar Player that Swarowski plans the market introduction of some limited edition Swarowski tubes? They are announced to have mouth blown envelopes made from true Bohemian crystal.

New ceramic metal eutectics of fibrous microstructure developed in the sixties by some Nigerian metallurgists (after the secret Nigerian space program had been abandoned) will work together with the mouth woven lattice of the Bohemian crystal in a way that some kind of crystalline clear and fully transparent sound can be created with these tubes that couldn’t be achieved before.

Now – as you know – tubes are electromechanical devices and the vacuum isn’t 100%. So the mechanical micro vibration of the envelop leads to a small amount of compression and decompression of the residual gas and by this affects the travel of the electrons fighting for survival.

After years of research the master glassblowers of Swarowski now did find a way to intentionally create a certain kind of resonance frequency for these mouth blown envelops by the way they blow. So customers now can custom order different envelops with quantum structures that depend on the individual vowel the glassblowers have formed with the mouth when blowing. So for the first time tubes now can be bought with pre-selected vowel like sounds (Ahhhh, Ohhhh etc.). The individual specs of the vowel (loudness, pitch, timbre, duration) are marked on the envelop (preamp tubes) and the marble base (power tubes) by Italian master calligraphers (preamp tubes) and chisellers (power tubes). So if the tubes have to be changed the same vowel-like character will be available again.

Cheers, have a nice Friday and a great weekend,

Max