Does it bother you that the picture is signed "ideal transient distortion"? This is a hint that the input signal is sent at a speed that is higher than allowed under normal operation. Open the Wikipedia article "Transient response". And all your distortion with overshoot at the start of the sine corresponds to the same measurements on the meander.
In normal operation, the rate of change of the input signal does not exceed the tolerance and no such distortion occurs. It looks like you need to read Jiri Dostal's book more carefully.
In the picture on the right, the signal is at 0 V before the 0 time mark. Then from that mark, it increases at a constant speed. Audio signals never change from one speed to another so abruptly, without transition.
Most people understand that a signal discontinuity such as a perfect square wave, can not exist in reality. A Voltage can not change from one value to another in zero time. Likewise, the derivative/slope of a signal nor the second (etc.) derivative can not change instantly. Hence, the drawing of an "Ideal transient" is not something that can exist. All systems have a bandwidth limitation and a delay between input and output. A delay is not a distortion because no information/harmonic frequencies are added to the signal.
Excellent point steveu
A good practical lab experiment of this is to record a 20 kHz square wave through a digital recorder with a 22 kHz BW and then play it back, looking at a scope. You get a 20 kHz sine wave.
A good practical lab experiment of this is to record a 20 kHz square wave through a digital recorder with a 22 kHz BW and then play it back, looking at a scope. You get a 20 kHz sine wave.
Unbelievable how this basic stuff keeps coming up time after time is various bizarre forms. A member, who appeared to be quite knowledgeable about speakers, was really struggling recently to understand why a woofer couldn't produce a 20Hz half cycle sine wave.
So it is OK if other member disrespect to me? Why do you use double standard?
In this forum so many amplifier's guru you can ask, Bob Cordell, Douglas Self, John Curl, Nelson Pass, Edmond Stuart, etc.
I am interested in relationship between the measurement and the sound. If you know something about it, I will learn from you.
A 20 Hz half cycle sine wave is not a sine wave. It contents harmonics.
A sine wave is a sustained* signal.
* Is sustained the right adjective ?
Last edited:
The reason people misunderstand is that human brains automatically produce explanations for sensory observations. By default the explanations are believed.
In this case an explanation should have been rejected due to conflict with theory. However, with no better and more plausible explanation readily available, the belief stuck. That's probably where it will stay until a more correct explanation for what is heard from different amplifiers is accepted. An explanation that the listener is imagining things will not suffice, and is probably not correct anyway.
In this case an explanation should have been rejected due to conflict with theory. However, with no better and more plausible explanation readily available, the belief stuck. That's probably where it will stay until a more correct explanation for what is heard from different amplifiers is accepted. An explanation that the listener is imagining things will not suffice, and is probably not correct anyway.
Which was where his problem started, even after all the explanations I'm not certain he understood. I struggled to understand why he didn't get it. He's very into phase correction with DSP and this could have clouded his thinking when it came to basic analogue signal processing, that's the impression I got.
My mistake it was about reproducing a single cycle and at up 200Hz, not that it makes any difference.
Last edited:
I agree , we can hear records with years of delay .
But what if this delay is not constant?
How big does it have to be for us to perceive it? We are talking about the biggest problems of audio amplification,
the first thing happens , the second thing is that it is very small.
I
Neither is a 20Hz full cycle sine wave. The whole concept of "first cycle distortion" is garbage.
I vote for "periodic".
@forr - yes, we never actually have real sinewaves as they start from -infinity and continue to +infinity. Any wave we tend to think of as a sinewave has a starting point (and an end) which adds harmonics. Mostly, after a few cycles the exponential decay of the initial transient has gone so for practical measurements a sinewave oscillator is virtually indistinguishable from a continuous wave.
Try putting a "sinewave" (from a standing start V=0 is a pulse really) through a simple RC filter.
You get the delay response. No active components were harmed in this demo.
You get the delay response. No active components were harmed in this demo.
Largely irrelevant. You could consider as a periodic signal and entire piece of music played in an infinite loop, so the period is something like 3 minutes. In this case, how relevant is the effect of a simple RC filter (and the first cycle distortion, of course)?
My point was really addressed to petr. Just to illustrate that you don't need an amplifier to generate the delay response he calls "first cycle distortion".
Otherwise, an RC filter is irrelevant (unless it cuts out significant audio frequencies).
Otherwise, an RC filter is irrelevant (unless it cuts out significant audio frequencies).
Hmmm... trying to point out that the first cycle distortion is a delay response seems to have got misinterpreted? RC filter generates a delay (are we agreed?) to an instantaneous wave change. Then a prop delay to quasi-continuous cycles at low frequencies (OK?) which turns into an increasing phase shift at higher frequencies as the RC network attenuates the signal.