What determines "believability of the reproduction illusion"
- Thread starter jkeny
- Start date
Hello Jkeny
Here is the definition
Your inner ear converts the analog waveform into discrete packets of information that are sent to your brain in the form of electrical impulses down the nerves. Your nerves do not fire continuously as they are chemically charged and are for all intents and purposes are sampling the waveform. They are encoded in both frequency and amplitude. So is it really that different?? Seems to meet the definition as hearing certainly is a form of information for your brain to process. I see that as food for thought and not in anyway shape or form as a black and white.
Rob
Rob,
As far as I have heard - I am not an expert is physiology - the retina is the only human organ that works in the digital mode. I hope our physicians can chime on this subject with more knowledge than me!
BTW, sampling and multiplexing are not forcefully digital operations - they can be performed in analog mode.
I just had a chance to partially come up to speed with this great thread, and I am still only up to page 12. To answer the original question about believability, and as I have posted a number of times before, I am in the same camp with JackD, Steve Williams, DaveC and others who put timbre at the top; my position continues to be that unless you get timbre right, you are not listening to music, but [distorted] sounds. Everything else is secondary, though very important, including the room. But when timbre is right, you recognize the instruments - thus the music becomes believable - from anywhere in the house (never mind just the audio room or the listening position), what you hear is highly real and believable, and I care less if an instrument is right in front of me or 10 ft away, or whether it lacks ultimate dynamics. This is not different than sitting anywhere in good-sounding hall - you enjoy the music from any distance to the orchestra and seat, because timbre is right, and things like presence and dynamics drift, or the "image" of orchestra may shift to your left or to your right (especially if you are sitting on the sides of a hall). So the fact I may be sitting on the sides of a hall does not retract from the sense of believability, because the image just happens to be shifted - big deal... I am still hearing the same instruments in all their timbral glory, and I know it's still the real thing.
To summarize: notions of dynamics, presence and tone are important, complete the picture, but they are just not the core and the root of it all - I find that timbre is. If the timbre is wrong, I feel you are just listening to the wrong thing, albeit with presence, color, dynamics, and all other secondary attributes. This is why I've called my system's page "Timbre & Articulation".
To summarize: notions of dynamics, presence and tone are important, complete the picture, but they are just not the core and the root of it all - I find that timbre is. If the timbre is wrong, I feel you are just listening to the wrong thing, albeit with presence, color, dynamics, and all other secondary attributes. This is why I've called my system's page "Timbre & Articulation".
Ack,
I think we mostly agree on the importance of timbre, in part because we are making an "audiophile" (this means vague and subjective ...) use of the word. Speaking in technical terms, how would you define timbre for the purpose of the debate in our great thread? I am not able to say from a recording if a piano is a Steinway or a Bosendorfer, but curiously I am able to say if sounds "believable" to me!
Timbre and resolution are also somewhat related, timbre becomes more accurate when fine details are present... but in other ways they aren't related as much. Timbre is the whole reason I use a silver/gold alloy instead of pure silver. Both have about the same resolution but the alloy makes acoustic instruments and vocals sound more realistic and believable.
You can't tell one piano manufacturer from another probably because you are not intimately familiar with their sounds, and not necessarily because your system can't reproduce timbre correctly.
I agree, especially that resolution improves timbre. Let me also clarify:
Articulation: all the information is there (instruments, voices)
Timbre: they all sound right
So my goal is to have all possible information reaching my ears and everything sounds timbraly correct. That's what's real to me.
Surely - I can not do it at even at concerts, unless I can see the plate ... But I could easily learn it in a few minutes, it is a question of interest.
But my ability is secondary to the thread - I would like to read how audiophiles define timbre - let us hope they do not say it is the quality that makes recordings "believable".
I've always thought of timbre as the character or quality of a sound that allows me to distinguish between the sound of a violin, a viola, a viola de Gamba, and a cello. It is also one of the characteristics (ingredients) along with dynamics, resolution, tonal balance, that allow me to believe that I am listening to a real, acoustic instrument rather than an artificial, plastic or amplified instrument. There are also cues like how "stringy" or "woody" the violin sounds, the quality of the attack, the transient, the decay, the tonal balance of the instrument. A system must preserve enough of the recorded signal and then reproduce it with enough resolution for one to claim that it is "accurate" in terms of timbre. That is, one must be able to hear a great deal of information and compare that with one's memory of past experiences with this sound to understand how a system conveys an instrument's timbre.
I also think one must hear accurate timbre for a system to sound believable.
In simple terms, timbre is the unique and characteristic sound of an instrument or voice, a property that enables us to distinguish sounds when everything else is equal (loudness, pitch, etc). In more complex terms, timbre is the collection of unique and complex characteristics, which include spectral (harmonics, subharmonics, inharmonics, etc - overall, the waveform), temporal (rise, decay), sound pressure, frequency and amplitude modulation, et al.
Hello ack
Agreed but there has to be a lot more to it than that as far as making something sound believable. After all you can easily distinguish differences between instruments and recognize singers voices on many meager systems. I can do that in my car. So even of you can get the timber correct enough to know who is singing or what instrument is playing that does not make it believable.
Rob
Hello ack
Agreed but there has to be a lot more to it than that as far as making something sound believable. After all you can easily distinguish differences between instruments and recognize singers voices on many meager systems. I can do that in my car. So even of you can get the timber correct enough to know who is singing or what instrument is playing that does not make it believable.
Rob
Notice, I simply defined a term. I also said earlier it is the most important aspect of believability, and we assume only IF and WHEN you get it right. Sure, we can all distinguish timbre in our cars and in meager systems (as you put it), but that doesn't mean they are believable, exactly because timbre - though different between instruments - is NOT accurate in such systems. So what we said is that timbre *has to be accurate* in order to make things believable... at least in the minds of some us...
Peter,
You are just addressing the classic definition of timbre - frequency response and decays, as I have referred before. However this definition is not enough for sound reproduction - almost all solid state electronics have excellent timbre if we just use these two aspects. Some people referred that electronics can change timbre, and give extraordinary importance to timbre. I think we need to know more about it. What is affecting our perception of timbre and is not timbre?
SET electronics systematically affects timbre, due to its output impedance. Curiously it is high on our "believability" scale.
I'm not sure what the definition of timbre is but I want to provide some examples of what I believe is often difficult for replay systems to reproduce convincingly.
The first example is applause. With a lot of systems, this can sound like rain on a tin roof & not like applause - is this an example of timbre?
The second example is the sound of brushing on drums (I don't know what that techniques is called - I'm sure there's a name for it?) - with a lot of systems this can sound just like hiss - is this an example of timbre?
The first example is applause. With a lot of systems, this can sound like rain on a tin roof & not like applause - is this an example of timbre?
The second example is the sound of brushing on drums (I don't know what that techniques is called - I'm sure there's a name for it?) - with a lot of systems this can sound just like hiss - is this an example of timbre?
SETs are high, in the right system, on believability because the nature of the distortions they make is sympathetic with the ear/brain's self-generated harmonics and masking that benefits low order distortions far more than high order distortions. Timbre is therefore perceived as more natural with a SET...a good one thta is. What do I mean by a good one? For starters it has to have a very well thought out robust power supply because there is no negative feedback to "clean-up" distortions from sloppy application of power. Secondly it MUST have a sufficiently larger transformer so that core saturation is not an issue up to at least the rated power. Surprisingly, this sin is committed by even very expensive SETs and Push/pull amps. Yes, Push/pull tube amps are just as gulity of these sins as well. Why do companies skimp? Good output iron is pretty expensive and usually the most expensive single component in a tube amp. It is also harder to get good high frequency extension from a large core output transformer. A good SET will have an output transformer that is low distortion in the bass and extension to well beyond 20Khz. If it doesn't do this, then the believablilty factor will be lower just like if the tube operational points are drifting around due to an inadequate power supply.
It's interesting that no-one here can really be too far from the truth because believability is ultimately at the core of this pursuit so I suppose getting to truth is inevitably tied up with believability.
Fidelity, transparency, resolution are the mainstays for a contextual truth and musicality, expressiveness and feeling are the truths of the spirit of music.
It also probably makes sense that a system good enough that it can truthfully portray timbre is perhaps likely to also be good enough at some level to also reveal dynamic shading, presence and express rhythmic structure and tonal colour.
Tonality surely lets us know that it is a real violin or cello that we are listening to but it is only with a believable level of these other elements that then let us know that it is Heifetz or Starker that are playing the instruments. All the elements that are critical to the artist are all the criteria that need to be brought to life to make us feel that this is truly believable. Tonality is vital for sure but equally it's in the dynamic range and shading and the vitality, nuance and presence that the artist evokes in their instruments that then directly speaks to us through their music.
Fidelity, transparency, resolution are the mainstays for a contextual truth and musicality, expressiveness and feeling are the truths of the spirit of music.
It also probably makes sense that a system good enough that it can truthfully portray timbre is perhaps likely to also be good enough at some level to also reveal dynamic shading, presence and express rhythmic structure and tonal colour.
Tonality surely lets us know that it is a real violin or cello that we are listening to but it is only with a believable level of these other elements that then let us know that it is Heifetz or Starker that are playing the instruments. All the elements that are critical to the artist are all the criteria that need to be brought to life to make us feel that this is truly believable. Tonality is vital for sure but equally it's in the dynamic range and shading and the vitality, nuance and presence that the artist evokes in their instruments that then directly speaks to us through their music.
An example I can give is the difference in timbre between two violins of different makers. You can, in real life, easily distinguish between a Stradivarius, and Amati and a Guarneri Del Jesu (I know as I have actually had the privelege of doing this right in my own home). It is the balance of the harmonics and overtones produced, their sustain and the way the balance of these harmonics changes with time. You can see this visually with a high resolution spectrum analyzer and a microphone.
I had my ex play the same note at the same subjective intensity (we stuck with a single note for the sake of being able to make sense of it on the spectrum analyzer) and we looked visually at the results. Sure enough, the spectrum of harmonics, while largely the same, had different weighting at different frequencies. All three violins sounded very different and also measured very different. What I couldn't do with the equipment I had was watch the decay of these harmonics to see if they all fell away at the same rate or if some "hung on" for longer...I am pretty sure that the balance changes during the decay and it is not the same with each intstrument.
To further muddy the waters, my ex-switched bows and the harmonic patterns changed again as did the sound, rather obviously.
Now, how does this relate to playback? I would wager that most systems would make this rather easy live exercise in identification of a distinctive "character" for each violin/bow combination into something rather more difficult. The first time I actually could tell by tone/timbre the difference between two oboes in an orchestral piece (the exact one escapes me now but I believe it was a very good recording of Stravinsky Petruschka) was with my Acoustat, Silvaweld, KR Audio setup. Before that an oboe sounded like an oboe and a clarinet like a clarinet but distinguishing an exact maker or era for an instrument would have been impossible.
A system that can resolve timbre to that degree is really taking a step towards believability and I heard maybe 2 systems at Munich (under show conditions mind you) that could possibly achieve this.
AH!! But you should be able to tell...assumming you are well versed in how each sounds live. Of course if you don't know what the real things sound like intimately then it is hopeless from a recording for sure.
A good recording of each played back by a top notch system should allow you to hear each instruments special tonal identifying characteristics.
I have had such systems where identifying different violins was possible.
