Oh - I already have access to this trick. I was confused by the reference to it being in the NEW issue of GENII. This effect can actually be found in the March issue of the magazine. Very nice it is too!
By the way - if anybody is interested in a brand new riffle shuffle principle, they should check out the following work from Leo Boudreau (who is one of the best thinkers in magic) - this was written up back in January of 2005...
www.themagiccafe.com/forums/viewtopic.p ... 6&forum=82
I have 'cut and pasted' the relevent information below:
A popular misconception is that a shuffle is a good way of mixing cards. The truth is that a shuffle hardly mixes them at all and you can prove it to yourself.
Take a pack of Zener cards and stack them cyclically in 123451234512345... order from the top down, where 1=circle, 2=cross, 3=lines, 4=square and 5=star. Now separate the pack between any star and circle, preferable near the middle of the pack, and shuffle the two portions together using a genuine riffle shuffle. Immediately after the shuffle and without cutting the cards, deal yourself five cards face-down one at a time into a pile. Then turn the fifth card over so it lies face-up on top of the pile.
If the shuffle were any good at mixing the cards, the fifth card would reveal little information about the rest of the cards in the pile. But the truth is that, when the fifth card is either a circle or a square, the pile will contain two circles, a cross, a square and a line card. When the fifth card is either a cross or a line card, the pile will contain two circles, two crosses and a line card. When the fifth card is a star, the pile will contain one card of each design. So, the fifth card by itself can reveal the identities of all the other designs in the pile even after a legitimate shuffle. Check it out for yourself.
Anyone care to explain this curiosity? It gets curiouser.
Suppose a line design sits on top of the first pile, a circle on top of the second pile, a star on top of the third pile, a line design on the fourth pile and a star on the fifth pile. These top cards together with two values (X+1 and 6-X) will help you to calculate the designs in all the piles.
X is simply the design on the pile immediately before the pile you wish to calculate. For example, if you wish to calculate the designs in the fourth pile, the value of the star on the third pile, which is 5, means X=5.
The fourth pile must contain the designs X+1, X+2, ... right up through the design sitting on top of the pile. The top card of the fourth pile is a line card, which has a value of 3. So the fourth pile must contain the designs 6, 7 and 3. Numbers larger than 5 are reduced by 5, so 6 becomes 1 and 7 becomes 2. Thus, the fourth pile has a circle, a cross and a line design.
When the number of designs obtained in the FIRST STEP is less than five, then the pile will contain additional designs 6-X, 7-X, 8-X, ... up to the number necessary to produce five designs. In our example, the fourth pile must contain a second circle (6-5=1) and a second cross (7-5=2).
Here is another example: If you wish to calculate the designs in the second pile, the value of the lines on the first pile, which is 3, means X=3.
The top card of the second pile is a circle, which has a value of 1. So the second pile must contain the designs 4, 5 and 6, the latter reducing to 1. The corresponding designs are the square, star and circle. Since this step has produced only three designs, you must proceed to the next step to capture two more.
The second pile must also contain lines (6-3=3) and a second square (7-3=4)
Needless to say, the process will also work for playing cards. Stack them cyclically in A2345678910JQK... order from the top down in the face-down deck. Separate the deck at the center, between the King and Ace, and hand the two halves to a spectator for a genuine riffle shuffle. After the shuffle, he will deal thirteen cards face-down into a first pile, thirteen cards face-down into a second pile, and so on.
The thirteenth card, the one sitting on top of each pile, will now serve to identify the cards (values, not suits). The two-step process is the same here as for the Zener pack, except that now the formulas are X+1 and 14-X.
Suppose a 7 were on top of the first pile and a King were on top of the fourth pile. Your calculations for the first pile would proceed as follows: X=13, X+1=1, and 14-X=1. (Remember, numbers greater than 13 must be reduced by 13.)
Based on X+1, X+2, ..., the first pile must contain the Ace through 7.
Based on 14-X, 15-X, ..., the first pile must also contain a separate set of cards running from Ace through 6.
The process will also work for the suits if they are stacked cyclically in CHSD order from the top down in the face-down deck. Separate the deck between a Diamond and a Club somewhere near the center and hand the two halves to a spectator for a riffle shuffle. After the shuffle, he will deal four cards face-down into a first pile, four cards face-down into a second pile, and so on.
The fourth card, the one sitting on top of each pile, will now serve to identify the suits. The two-step process is the same, except that now the formulas are X+1 and 5-X.
Here is another riffle shuffle effect you may enjoy doing. It's called:
You produce two decks of playing cards, separate one of the decks at its center and hand the two halves to a spectator for a genuine riffle shuffle. You keep the other deck which, you claim, has been arranged based on a dream you had the night before. Without shuffling your deck, you place it in your pocket and begin pulling cards out of your pocket and dealing them into a pile, turning them face-up one at a time. The spectator follows your lead, dealing the cards from his shuffled deck into a pile and turning face-up one at a time. Notice that you remove the cards first, followed by the spectator. Oddly enough, the cards dealt from both decks match perfectly, card after card.
The first deck, deck A, is the one that will eventually go to the spectator. The second deck, deck B, is the one you will keep. Deck A must have a one-way back design. Shuffle it thoroughly so that the cards are truly in a random order. Then orient the backs of the first twenty-six cards so they all point one way and orient the bottom twenty-six cards so they all point the opposite way. Deck A is now ready.
Deck B is prepared next. Arrange it so it is in the exact same order as deck A. Then reverse the order of the twenty-six cards at the bottom of deck B. When done, the first twenty-six cards in both decks should be in the same order and the bottom twenty-six cards should be in reversed sequential order. Deck B is now ready. It will end up in your pocket.
Separate deck A at the point where the back design changes from one orientation to the other. Hand both halves to the spectator for a genuine riffle shuffle. No cutting is permitted either before or after the shuffle unless the cut is false.
You will glance at the top card of the shuffled deck to see how it is oriented. If its orientation indicates that the card came from the top half of the spectator's deck, then remove the top card from deck B in your pocket and drop it face-up on the table. If, on the other hand, its orientation indicates that it came from the bottom half of his deck, then remove the bottom card from your deck. In either case, the spectator will then deal the top card from his deck face-up. The cards will match.
Note how the back of the next card in the spectator's deck is oriented. Use it to determine whether you should deal from the top or bottom of your deck. You will deal the appropriate card and the spectator will follow you by dealing his top card. Continue dealing in this manner and the cards in both decks will match perfectly. Keep the cards coming from your pocket as rapidly as you can and slap them down on the table.
All the best, and congratulations to Leo on another amazing discovery!
PS. If Max Maven should read this - he mentioned he had a solution to a riffle shuffle problem (using cards with letters printed on the back - originally a Martin Gardner trick called 'SHUFFLED') in the bound reprint of AZIZ magazine (that was released along with IBIDEM Vol.3). I was just wondering if he would maybe release his solution in his future work on The Gilbreath Principle? I have worked on it myself, and would love to see what Max came up with as well...