cable directionality


I've always heard that single-ended, directional interconnects (e.g., AudioQuests) should be installed with the arrows pointing away from the source. Is that really true, though? It seems like it actually doesn't have much to do with signal travel, but with grounding: normally, the shield is connected to ground at the end the arrow points toward. Therefore, it seems that the cables should actually be installed with the arrows pointing toward the system ground (normally the preamp). This makes a difference (or does it?) for the cables between the preamp and power amp. What about directional speaker cables? Should the arrows point back toward the preamp, too?
ezside
You are absolutely right re: arrow points to ground, with some notable exceptions (Magnan and Purist being two I'm aware of). The general rule (for IC's) is all arrows point to the preamp, which should also be the only component whose ground (third AC plug pin) is connected to ground at the wall.

With SC's though, it doesn't generally matter (there's no ground or shield) except that the dielectric will take a "set" after a while, so unless you want to have to break them in all over again, it's best to follow some protocol and always install them the same way.
AudioQuest ICs, "The plug with a single printed band at the sending end and the plug with the double printed band at the receiving end". > goes in the direction of signal flow. CDP > Preamp > Power Amp. This is the correct way to install the Lapis and Diamond ICs. Below is some info on AQ design.

http://www.avalon.co.nz/audioquest/aq_theory5.html

http://www.positive-feedback.com/Issue18/audioquestinterview.htm

Oh boy! There is some confusion here. I always heard that the arrow points in the direction of signal transmission (preamp to amp), which means that, for conventional grounding practice, the IC shield would be grounded at the TAIL of the arrow. I use my ICs this way. However, depending on your particular electronics, it may make no difference, or be better the oposite way around. Grounding is not an exact science, and a little experimentation is appropriate.