Actually, using an .mp3 to shoot out between mics is a bit unfair. Here's why; most .mp3 codecs if not all, have a upper frequency response somewhere shy of 20kHz, if I recall correctly. Note that the SM57 response is greater than 0dB, meaning a boost, from about 2 kHz all the way well past 10 kHz, whereas, the Rode is nearly flat UNTIL about 14 kHz, where it then shows a bump. The SM57 is greater than 0dB from about 2 kHz all the way up to 15-16 kHz (difficult to read correctly on the log plot - but visible in greater detail on the .pdf: http://www.shure.com/idc/groups/public/documents/webcontent/us_pro_sm57_specsheet.pdf )

This would make the SM57 seem 'hyped' in the high end of the available spectrum in the .mp3 file compared to a more flat response of the Rode. The SM rolls off below 200 Hz, but most people high-pass recordings for vox between 100 and 200 Hz, so this wouldn't be necessarily apparent.

Also take into account that many people have some age-related loss in the region where the two mics differ, particularly where the RODE shows a non-flat response in the teen kHz region, though it's pretty doggoned flat for a music microphone. You don't see much more flat response than that except with measurement microphones.

Regardless, trying to do a shootout with out simultaneous recording at each mic's optimal position makes this kind of an exercise difficult, and lossy compression throws another curve into the situation that makes it difficult.

Is it something to worry about - not at all.

-Scott