I am currently running composite directly from a CLD-59 LaserDisc player (I'm looking to get s CLD-99 when the budget allows) to a Mitsu WS65511 display. My understanding is that on LD, the signal is composite to begin with, so that it is better to run it directly to the display, and use it's 3-D Motion Adaptive comb filter instead (the S-Video uses the CLD-59's 3-line filter). My question is whether it would yield a better picture by running composite to a Denon-2105 and letting it split the signal into component? I know that it will not add anything not already there, and does not do upscaling, but I was thinking that splitting the signal to seperate Y, B, and R components might yield a cleaner picture. Does anyone have any experience with this?