It may be time for me to jump in here. While I apprecaite all the comments, and empathize with those who love audio and the attempt to explain what they hear though technical terms, there comes a point where it makes sense for the designer to make a stab at clearing the air. I will address two issues: 1) Relevant factual information that might shed some light, and 2) Technical information intended to both enlighten and confound, strictky due to the real world complexity of why things sound as they do.
1) The Italian article did a very good job of explaining to a general audience the difference between "velocity" devices and "displacement" devices, and how RIAA is handled. That was, in part, their objective - it was to inform.
2) Comments that attempt to quantize one's experience by referring to amplitude pertubations are equivalent to four blind men each describing what an elephant is by touching varied parts - and invariably giving an incomplete view of the animal. Sound is complex - yes - if there is a terrible amplitude anomoly, it makes for bad or unlistenable products. But audio has a long litany of products that are quite wonderful that are not "flat".
The human ear lives in the time domain - if this fact is not understood by enthusiasts approaching the field, it is part of audio 101. There are those who insist on .1dB flatness without undertanding that a full sytem tuned and maximized for flatness often introduces phase or time error that are a cure worse than the disease. The small pertubation bumps one sees on a speaker reponse are just that - phase interractions. They give, in part, the tonal flavor of a product.
Although we have made some recent changes on our preamp design to realize a very small change in amplitude and phase flatness, the original product was well received by most listeners and owners. If the frequency response were very bad, that would not have been the case.
What is important to gather (which is hard to do)is what time errors occur in any product, how many, where do they exists in the freq domain, how far do they spread, and how serious are they? This describes only one part of the problem when trying to use technical terms to exlain what we hear. We listen to speakers that are far from flat, in environments that are far from perfect, and often get very good results. Why is that??
The answer in part is that. What is the rise time of the product (how fast is it). Does it ring (resonate) if it is a transducer. How much time shift is there, and how much. THEN, what is the overall response.
If a products design criteria are in accord with certain critical parameters that physics say are good things, it may in fact work well. From then on, your ears are the judge. But do get the laws straight. Amplitude flatness can easily be done at the sacrifice of other MORE IMPORTANT parameters.
Peter Ledermann/President/Soundsmith
1) The Italian article did a very good job of explaining to a general audience the difference between "velocity" devices and "displacement" devices, and how RIAA is handled. That was, in part, their objective - it was to inform.
2) Comments that attempt to quantize one's experience by referring to amplitude pertubations are equivalent to four blind men each describing what an elephant is by touching varied parts - and invariably giving an incomplete view of the animal. Sound is complex - yes - if there is a terrible amplitude anomoly, it makes for bad or unlistenable products. But audio has a long litany of products that are quite wonderful that are not "flat".
The human ear lives in the time domain - if this fact is not understood by enthusiasts approaching the field, it is part of audio 101. There are those who insist on .1dB flatness without undertanding that a full sytem tuned and maximized for flatness often introduces phase or time error that are a cure worse than the disease. The small pertubation bumps one sees on a speaker reponse are just that - phase interractions. They give, in part, the tonal flavor of a product.
Although we have made some recent changes on our preamp design to realize a very small change in amplitude and phase flatness, the original product was well received by most listeners and owners. If the frequency response were very bad, that would not have been the case.
What is important to gather (which is hard to do)is what time errors occur in any product, how many, where do they exists in the freq domain, how far do they spread, and how serious are they? This describes only one part of the problem when trying to use technical terms to exlain what we hear. We listen to speakers that are far from flat, in environments that are far from perfect, and often get very good results. Why is that??
The answer in part is that. What is the rise time of the product (how fast is it). Does it ring (resonate) if it is a transducer. How much time shift is there, and how much. THEN, what is the overall response.
If a products design criteria are in accord with certain critical parameters that physics say are good things, it may in fact work well. From then on, your ears are the judge. But do get the laws straight. Amplitude flatness can easily be done at the sacrifice of other MORE IMPORTANT parameters.
Peter Ledermann/President/Soundsmith