XLR or single ended cables and why?

Shadorne - If implemented properly a balanced component uses differential circuitry throughout - each phase (+ and -) of the signal has its own signal path. There are pros and cons to this, but it is far more than just another connector type.

There are lots of other threads that explain this in gory detail.

Balanced audio interconnects (XLR) utilize two conductors, while standard RCA audio interconnects use only a single "unbalanced" conductor. This enhanced design gives balanced audio interconnects the power to provide greater resistance against sources of noise (RF, EMI, etc.). The result is a more pure signal transfer, especially in longer-run applications.

You may or may not hear a difference between XLR and RCA in short runs, however XLR should prove quieter on very long runs or in set-ups where many cords come in contact with each other...think rock concerts, where all instrument and mic cords are XLR.

Tvad sez:

however XLR should prove quieter on very long runs or in set-ups where many cords come in contact with each other.

Exactly. Plus the common mode rejection (i.e. "noise" that is common to both sides of the channel is cancelled out). Often it is twice as loud (spl) at the same volume control setting.

Shadorne, Is it a big deal to have an extra increase of 6db with XLR ? That's a lot of extra juice. Running either at 4 or 8 ohm, It requires double in wattage power for each 3db increase. Isn't it ? BTW, I have my MD. pulled the wool out of my ears the other day. Thanks for the advise.

Simple answer: With your Krell gear, you should definitely try out a balanced connection (and this is regardless of how long your cable run is). You don't need to commit first to purchasing the XLR interconnects if you can borrow a pair from a friend or a store demo for audition. Of course it would be ideal if you could either get ahold of a pair that corresponds to your usual reference single-ended IC's, or else wrangle pairs of the same audition model in both XLR and RCA versions so that the comparison isn't confounded by uncontrolled variables. Then just listen, not for noise per se, but for probably somewhat subtle (which is not to say unimportant) musical differences, especially concerning transient, dynamic, spatial, tone color saturation, and transparency/resolution qualities. Remember that because a balanced connection will usually yield a higher volume than single-ended (typically about 6dB hotter), for valid results always compensate with the volume control to equalize levels when performing A/B comparisons.