recommends for interconnects?

I had MIT Shotgun cables for many years.  I replaced them a few years ago with Nordost cables, and the change was dramatic.  Audio stores and manufacturers usually hook their components up with Nordost cables to show off the benefits of their components.  

@jafox is correct, you need to buy cables that make a difference.  For Nordost, Heimdall is the tipping point where everything at that level and above makes a big difference.  Cables are as important as components.  Look for used cables, they have a nice discount.  I've upgraded most of cables to Nordost Valhalla or Odin, but I got them used.  I went to an audio store a couple days ago to listen to Sonus Faber Olympica Nova V speakers, and they had them hooked up with Nordost Valhalla cables.

Cables make no objective difference. You said it in your own post, "..so many opinions" and that's all they are without hard evidence to back them uo. Save your money, cables NEVER get "tired". 

@mclinnguy In useing xlr cables with Nortrick ends in the professional world they only wear out from around at the connectors from being coiled and packed up for the next show or someone runs over the cable with a heavy object that possibly crushes it that's why the band gear usually gets set up last on stage to prevent this from happening.. Sure others will comment but thes are the cables we use for touring that get connected into the splitter which seperater the signal to the monitor board and also send signal down a fiber line to the FOH mix.

Beldon Wire Iconoclast.  

A taste of the very high end at reasonable prices with good service... and you can try them and return if not satisfied. So, no risk.

The reason is interesting.  SoulNote says music is a waveform delivered in time, not a Fourier transform delivered in the frequency domain. Almost all cable measurements focus on the frequency domain.

Why no shielding?  Well, the cables are balanced so EMI should equally affect the + and -, cancelling out.  Shielding increases capacitance, which smears the signal in the time domain.

@richardbrand spot on!