Here is a description of the Devant effect. Consider the directness (David Devants' method in "Lessons in Conjuring):
One spectator is invited onstage to assist the magician. The spectator cuts a shuffled deck into three piles and selects one pile. He then counts the cards in that selected pile, slowly and accurately, by dealing them back to the table singly and from a hieght of about 12 inches above the table. After the deal, the spectator gathers the pile and puts the cards in his inside jacket pocket and buttons his jacket. The performer gathers the remaining cards and, stepping into the audience, has a second spectator select any card, the number of pips on the selection determining the number of cards to go across. So if a three is selected, three cards are passed from the deck in the performers' hands to the packet the first spectator is holding. After pantomiming passing the cards the performer asks the spectator to remove the cards from his pocket and count them, one at a time, showing that the requisite number of cards has indeed arrived!
Note the economy of motion - the lack of repetitive counting, the appearance of the spectators doing everything, the performer never touches the cards - it seems as if the performer simply directs the action - this is really a beautiful version. There is only one spectator onstage, it is much less potentially confusing/distracting from the audience viewpoint.
Short of quoting the original text, I have given you all you need to figure out the method. Furthermore, the appearance is the part that matters, and believe me, what you just read is exactly how this appears to the spectators.
By the way, "Lessons in Conjuring" is NOT all that hard a book to find used, and not particularly expensive - if you are really interested try H&R. Happy hunting!