音响的终极目标是追求动态？ [复制链接]

LESLIE，我觉得你对“瞬态”的解释是对的。但是，你的解释也恰恰说明了“瞬态”不是“动态”。
“瞬态”是声音的动态响应的一种“形状”，与电平/信号/声压大小无关。因此，我们往往会对信号波形的前、后沿进行分析来推断它的“瞬态”好不好。对一个持续的弱信号，即变化不大的信号，我们也可以分析它的“瞬态”，但是因为它的动态很小，分析动态就没有意义。反之，一个持续的强信号，也可以分析“瞬态”，但是它虽然信号强大，但是无变化，或者变化很小，分析其动态也就没有必要。因此，虽然”瞬态“和”动态“都与时间有密切关联，但是，它们是两码事，这就是我对“瞬态”和“动态”区别的理解。

看到现在，我不禁回过去再看一遍MENN兄的两篇贴文。我觉得MENN兄写出来的情况是非常真实的，是每个人都会碰到的。现在本人的理解是：
1）动态范围：信号/声压的最强值和最弱值之差，和时间没有关系，但是和系统的本底噪声或系统的信噪比有非常密切的关系。不谈信噪比，而去追求动态范围是毫无意义的。在不考虑别的指标时，动态范围越大越好。
2）动态：是完成信号/声压的最强值和最弱值的变化时相对时间的比例，或者说需要多长时间，系统才能够完成“动态范围”指标中的信号/声压变化。因此，它和时间有密切的关系。离开了对时间的评估，动态也就没有意义。和动态范围一样，在不考虑其他指标时，动态越大越好。
再从自己发烧的体会来看，要把整个系统的动态提高、增加动态范围真是一件非常非常难的题目。不论你是听音乐还是看DVD影片都是如此。

供参考，对“动态范围”又一解释：

DYNAMIC RANGE

Electroacoustics / Acoustics
The difference in SOUND PRESSURE LEVEL between the saturation or overload level and the BACKGROUND NOISE level of an acoustic or electroacoustic system, measured in DECIBELs. This RANGE may be expressed as a SIGNAL-TO-NOISE RATIO for maximum output. Compare: GAIN.
For a sound or a signal, its dynamic range is the difference between the loudest and quietest portions. Compare: LOUDNESS, POWER, VOLUME.
The human hearing system has a dynamic range of about 120 dB between the THRESHOLD OF HEARING and THRESHOLD OF PAIN. See: RECRUITMENT.
The dynamic range of music may approach 90 dB, and since the maximum range is 55-60 dB for analog tape, and often less for disc recordings and optical soundtracks, COMPRESSION is necessary in the recording of such signals. Noise reduction and companding equipment are designed to increase the dynamic range by 10-20 dB or more. The dynamic range of DIGITAL RECORDING is generally over 90 dB.
See also: LIMITER, LINEAR, SOUND SYNTHESIS, VU METER. Compare: DYNAMICS, SILENCE.

供参考，对“动态”的一种解释。

DYNAMICS or DYNAMIC

Music / Electroacoustics
Strictly speaking, dynamics refer to the variations in LOUDNESS of a musical composition or specific NOTEs.
Compare: DYNAMIC RANGE, VOLUME.
The most common dynamic markings, from quietest to loudest, are the following:
pp    pianissimo (very soft)
p     piano (soft)
mp     mezzo-piano (medium soft)
mf    mezzo-forte (medium loud)
f    forte (loud)
ff    fortissimo (very loud)
These terms have no absolute values and are relative to one another according to the context of the music. Changes in dynamic levels are indicated as follows:
cresc. crescendo (increasing loudness)

decresc. decrescendo (decreasing loudness)
In ELECTROACOUSTIC applications, the term dynamic refers to time-dependent behaviour, such as the motion of a dynamic MICROPHONE, or the time-dependence of dynamic filtering or spatial modulation.

amrs 在 2003-8-13 17:49:21 发表的内容 由衷地,我希望有条件的发烧友去听一听LP,特别是走在摸拟最前端的最新现代模拟lp之声,无论cd,sacd在规格上是多么的优异,反映在声音听觉上,与我运用lyra tatin或clear audio 参考唱头模拟组合在逼真还原录音音乐演奏实像及音响直线跟随音乐动作变化的再现感,均不在一个数量级别内,重播大编制音乐lp具压倒性优势! 虽然,或有人认为我狂妄或是耳障人士,但我依然希望传达给更多人知道.

Amrs兄，我对LP具有更大动态还是想不通。LP的母版是通过刻纹刀来记录的。因为LP是模拟信号，因此LP没有数码产品的A/D、 D/A两次转换的缺点，因此它的声音还原的象真性很好。但是它的纹路横向的振幅受到相邻纹路距离的限制，否则就会“跳槽”。因此，常常会先用“动态压缩”的方法减小它的振幅，然后在放音时用“动态扩张”的方法还原。它怎么可能做到很大的动态呢？

供参考：
Dynamic Range
The dynamic range is often represented as a log ratio of well depth to the readout noise in decibels. For example, a system with a well depth of 85,000 electrons and a readout noise of 12 electrons would have a dynamic range = 20 log (85,000/12), or 77dB. The higher the number the better.
This ratio also gives an indication of the number of digitization levels that might be appropriate for a given sensor. A sensor with a well depth of 35,000 electrons and with 15 electrons of read noise would yield a ratio of 35,000/15=2333. Unless there is a desire to resolve this noise with a 16 bit system into 65,536 gray shades, a 12 bit system with 4096 levels would be more than adequate for this sensor. In this case, about 2 counts of noise would be digitized with the signal. For a KAF-1300, which has a well depth of 150,000 electrons and 15 electrons read noise, the ratio becomes 10,000, or 80dB. This sensor would be most suited for 14 bit digitization (16384 levels). For the SITe-based AP7, with a well depth of 350,000 electrons and 15 electrons of noise, the ratio becomes 23,333, or 87dB. Digitization to 16 bits is appropriate in this case. The typical values achieved for the AP7b with the SITe CCD is a noise of 9e- and a digitized well depth of 300Ke-.   This yields a dynamic range of 33.3K:1 or 90.4dB, OR just over 15 bits.

ERIC，谢谢回答。其实，发烧友都是很贪心的 - 手里拿着个60米，眼里会瞄着个100米，最好还是个只跑9秒的。

Eric,仍然拿“A君要2分钟才能跑完100米，B君10秒就跑完60米”来做例子，是否就说明：“动态大的动态范围不一定大，动态范围大的动态不一定大？”
因为2分钟和10秒的差别太大，如果假设A君要12秒跑完100米，B君要10秒跑完60米，哪个好听呢？