Oh boy, BIG can of worms here, this is a large complex subject. I'll try and distill it into something short, unfortunately that means simplifications etc.
Anyway this is all about what is called transmission line theory, how signals travel down whats called a transmission line, a coax cable is a transmission line (I'm NOT going to define a transmission line here, go look it up if you really want to know) The gist of all the math and books and books on the subject is that as long as the source impedance, the transmission line impedance and the load impedance are all the same, a pulse fed into the transmission line (TL) will come out the same as it went in, but diminished in amplitude. (note the actual value is not important, as long as they all match) If the impedances do NOT match strange things happen. The primary impact is that when a pulse hits an impedance mismatch some of the energy in the pulse gets reflected back the way it came. The pulse that continues on is diminished in amplitude (conservation of energy) and the reflection is opposite in polarity. An example, lets say a 1V positive going pulse hits the mismatch, lets say the mismatch is such that 50% keeps going and 50% gets reflected, thats a 0.7V positive going pulse continuing, and a negative going 0.7V pulse reflected back. If the source to TL impedance match is perfect nothing else happens, the reflected pulse gets absorbed. BUT if there is a mismatch at the source (such as when using a 75 ohm cable with RCAs at both ends), it gets reflected AGAIN. So now we have a positive going pulse of 0.5V going down the TL with a delay equal to twice the delay of the length of cable. When it hits the load mismatch it gets reflected again etc. So each original pulse results in a series of pulses with decreasing amplitudes showing up at the receiver. What happens at the receiver is very dependent on the cable length. A positive might hit a positive, increasing it's amplitude, a positive might hit the negative of another pulse, canceling it out, etc. etc. The result is that instead of a nice clean square wave you will get some very bizarre looking waveforms with a stair step looking profile. Changes in length of the cable can completely change the look of these waveforms. What is a poor receiver to do when it gets these massively messed up waveforms? Amazingly pretty good. Modern circuits can take this garbled looking mess and most of the time extract something usable out of it. BUT frequently the recovered signal can have some significant amount of extra jitter (with interesting spectrum) because of this. There is also a significantly strange pattern of noise injected into the power supply and ground plane for any real implementation of a receiver getting one of these signals. Note that the exact cable length and amount of the mismatch can significantly change what the signal looks like at the receiver. And to make it really fun, very few sources and receivers really are 75 ohms, it ranges from about 20 ohms to 120 ohms or so depending on exactly what parts are used and the precise details of the board layout. Only a very tiny percent of designs actually lay out the board calculating or measuring the impedances of the board traces. I frequently wind up laughing hysterically when I look at pictures of expensive designs and see how far off their board design really is. The upshot is there is no such thing as the "one best cable", it is a massive crap shoot. To a large extent it depends on how well a particular cable just happens to match the impedance of your source and receiver rather than any inherent "goodness" of the cable. Even with exactly the same cable changing the length by 2 inches can completely change the behavior. So what is a user supposed to do? The only real answer is hedge your bets as well as you can, use equipment with real 75 ohm connectors, use cables with real 75 ohm connectors, and then try lots of different cables from different manufacturers. This brings up the issue of expensive audiophile cables. Some of these may actually be slightly better cables than well made ones from say Blue Jeans, BUT you still have the crap shoot. Just buying an expensive cable is NOT going to guarantee it is a good match for YOUR system. A well matched $5 cable is going to sound better than a not so well matched $200 cable. You might get lucky and the $200 cable is the best match for your system, but it can get awfully expensive trying lots of different expensive cables to find the one that matches YOUR system the best. That is why MY approach is to try lots of different inexpensive cables of different lengths and stick with the one that sounds the best. Well actually my approach is to make everything myself so I KNOW it's well matched to begin with, but that's a little out of the realm of possibility for most people. So there you have it, far more than you ever wanted to know about transmission lines and digital audio. John S. ------------------------------------------------------------------------ JohnSwenson's Profile: http://forums.slimdevices.com/member.php?userid=5974 View this thread: http://forums.slimdevices.com/showthread.php?t=97881 _______________________________________________ discuss mailing list discuss@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/discuss
