UID829861
威望0
金钱3648
交易诚信度0
主题0
帖子249
注册时间2011-6-17
最后登录2017-2-22
初级会员
 
交易诚信度0
注册时间2011-6-17
|
一个外部时钟反对派的看法和对Lavry看法的解释:
http://www.gearslutz.com/board/3663212-post39.html
大概的意思就是内部时钟已经做的足够好了,要求器材去和外部时钟保持一致在工程上是一个很难做到的要求,一个器材若果连内部时钟都达不到要求,那么希望他能和一个高质量外部时钟保持一致是不太现实的。
I think, to properly evaluate those claims, you should look into just how clocking fits into AD conversion as well as precisely what happens when a converter is asked to slave to an incoming clock source.
One thing that is somewhat crucial to understand: the clock signal is just that. There is no magic voodoo carrier wave built into it. It is simply a clock signal that is supplied out to devices which must slave to a master. It is either good, accurate clock. Or not.
As discussed above and in the aforementioned thread with converter design legend Dan Lavry, the job of a converter's internal clock is relatively simple: it uses a crystal controlled clock circuit to provide highly regular sample timing x thousand times a second.
When we ask a converter to slave to an external clock source, the converter uses a phase-locked loop to attempt to maintain sync with the incoming signal. This is a much more difficult job than simply keeping time. The PLL must, in effect analyze the incoming clock signal and then continually modify its internal clock timing to attempt to stay in sync with the incoming clock.
While it's possible (almost anything is possible in conjecture-land) that a converter might have a really crappy basic clock circuit and yet still have really excellentPLL/slaving circuitry so as to benefit from external clocking, it's probably highly unlikely, particularly in modern, professional gear.
The most likely result of this continual attempt to adjust internal timing ot outside clock is increased jitter -- an increased irregularity in the timing of sample firing within the converter, resulting in (possibly) audible sample timing slur.
While it's possible that some will actually prefer this form of distortion, a good read of the Lavry posts in the above-referenced thread will reveal that the nature of that form of jitter distortion is a far cry from the 'good distortion' some of us like to get from overdriving analog amplification circuits or saturating tape.
Of course, when you have multiple converters running in tandem, there has to be a single master clock source. It can be supplied by one of the converters, or from an external source. And it may well be convenient in a larger rig to use a central, external clock source with multiple routing possibilities. |
|