感谢YING兄推荐好文,我翻译出来给大家讨论,有什么错误请大家指出。
Signal-to-Noise Ratio, Headroom and Dynamic Range
There are a couple of related concepts we need to consider here. Dynamic range is the total range of amplitudes a system can handle. For our ears, that dynamic range is 120 dB from threshold of hearing to threshold of discomfort. For audio systems dynamic range can be defined as the range of amplitudes between the noise floor of the system and its onset of clipping (the level at which the power supply is no longer adequate to provide larger waveforms). For acoustic spaces, the dynamic range can be defined as the range of sound pressure levels between the acoustical noise floor of the space and the onset of non-linearity in the air (around 130 dB SPL).
有几个容易混淆的名词在这里解释一下。动态范围是指系统能够处理的总的振幅范围。对于人耳,动态范围达到120dB时,达到人耳痛阈。音频系统的动态范围可解释为从系统底噪到信号刚刚进入削波状态(电平处在电源供应已不能充分满足较大波形状态)的振幅范围。在声学控制室,动态范围可以被解释为从从空间的声学底噪到在空气中刚刚出现非线性传播的声音的声压级的范围(130dB SPL左右)。
For convenience, we often divide dynamic range into two more specific ranges: signal-to-noise ratio and headroom. Signal-to-noise ratio is the range between some nominal maximum signal, which we usually call 0 VU or –14 dBFS, and the noise floor of the system. Headroom is the range above that nominal maximum signal to the onset of clipping (er, distortion). So, headroom plus signal-to-noise ratio equals dynamic range.
为了方便,我们经常把动态范围分成两个更小的范围:信噪比和headroom。信噪比是指在一些额定信号,就是我们常说的0VU或-14dBFS,和系统底噪之间的范围。headroom是指在额定信号之上,到刚好进入削波状态(失真)的范围。净空间加上信噪比等于动态范围。
While this is obvious, it has some important ramifications that are easy to forget. If we increase headroom, we decrease signal-to-noise ratio. If we improve signal-to-noise, we decrease headroom. The two are related and always limited by the overall dynamic range of the system.
虽然很明显了,但还有一些重要的细节容易被忘记。如果我们增加Headroom,就得减小信噪比。如果我们改善信噪比,就得减小Headroom。它两个被系统的总体动态范围所关联和制约。
Take a look at some relevant dynamic ranges that we have to work with in audio: the range of our hearing, the range of a “good” acoustic environment and the range of a “good” analog audio signal prior to recording. Note that the limiting factor here turns out to be the acoustic environment. Interesting, eh? In fact, the dynamic range of a digital audio recorder, which we are not discussing here, has about the same range as the acoustic environment, ca. 90 dB. Another observation is that the actual range of the acoustic noise floor to the loudest acoustic sound is probably going to be substantially less than 90 dB SPL. Hardly anybody plays at 130 dB SPL!
检查一下我们工作中不得不遇到的一些与音频相关的动态范围:听力的范围,“好”声学环境的范围,和录音前“好”模拟音频信号的范围。记下限制它们的因素,以弄清声学环境。事实上,我们在这里不细细讨论的数字音频录音机有着和声学环境差不多的动态范围。Ca. 90dB。另外,声学底噪达到自然声响的最大值的实际范围的测量在一般情况下,会完全小于90dB SPL。几乎没有人能播放到130dB SPL。
Another good point to keep in mind is that the overall, final dynamic range of our recording will usually be determined by the smallest headroom range and the smallest signal-to-noise ratio the signal encounters on its odyssey from air to electricity to ear. So in Figure 1 above, the dynamic range of the signal will be 90 dB in spite of the fact that the audio signal and our hearing both have ranges greater than that.
其他需要记住的好点子是全面的。我们记录的最终动态范围是由信号从空气到设备再到耳朵中间遭遇的最小的信噪比和Headroom所决定的。所以在上图1,信号的动态范围将有90dB,尽管事实上音频信号和听力的动态范围都比它好。
这篇文章清楚的解释了动态范围、信噪比以及headroom(净空间)之间的关系以及在工作中的应用和需要注意的方方面面。信噪比和Headroom都属于动态范围的一个部分,而信噪比和Headroom又相互制约。而对于文章中的图更是值得研究。希望此文会对大家有所帮助。
[
本帖最后由 margindge 于 2006-8-13 21:57 编辑 ]
本帖子中包含更多资源
您需要 登录 才可以下载或查看,没有帐号?注册
x