Generally speaking, many tapers use hypers/cards/omnis as follows:
- omnis - provide a very open, natural sound, with no real attenuation of the surrounding environment; typically used on-stage or fairly close to the sound source in a good sounding environment in which one is not concerned about the environment (crappy venue acoustics, loud audience, etc.)
- cards - kind of the all-purpose pattern for many tapers, commonly used in a variety of situations (on-stage, FOB, from the "tapers section", etc.); not as open and natural a sound as omnis, but not as closed or colored sounding as hypers; the pattern helps reject some of the crappy room acoustics and / or obnoxious audience; many tapers start off with (or use only) cards because they're quite flexible, providing some of the pros (as well as some of the cons) of both omnis and hypers, and they may not have the budget to buy multiple sets of mics or caps to include multiple patterns in their mic arsenal
- hypers - better off-axis rejection than cards, but not as open or natural sounding as cards (which in turn are not as open or natural sounding as omnis); due to the off-axis rejection, often used to help minimize an unpleasant environment (e.g. crappy venue acoustics or obnoxious crowd) or less than ideal location (too far back, hypers will help the recording sound closer to the source);
Keep in mind that even within each mic pattern, different configurations may yield significantly different results. For example, cards ORTF may sound noticeably different than simply pointing the mics at the stacks. All that said, it really does depend on what one's own ears like to hear - different people have different tolerances for how much crowd noise is irritating on playback, or different preferences for the balance between openness and naturalness one likes to hear in a recording v. the crowd noise and/or venue acoustics.
In some ways, I think db and Teddy are both correct: physics essentially determines the results based on environment, pattern, configuration, etc., and many people prefer similar sonic characteristics in their recordings (db), all else being equal (e.g. environment) different people may (and often do) prefer different results as a function of different mics, patterns, configs, etc. (Teddy).