I use 44.1 so no down sampling is required. I am a little curious though as to what issues down sampling might cause. Is there a reason to go with even multiples of 44.1 instead of the maximum possible such as 96 or 192? Computers are designed to do complicated math so I would think this should not matter.
I clearly hear a nicer top end at 88/96. However I don't have a dual processor on my computer yet so I am a little bit limited to how many plug ins I can use on a song. Waves and some UAD plug ins also are optimized at the higher sample rates.
PCM 701 (emp. off)->PCM601->2xDTC77 (for simul. rec. or dat clone)->CDR620 (gold on gold archivedisc).
And for restauration (not very often) the FC1 by ADT with pc harddisceditor.
Standalone, or in combination.