Anonymous user
Talk:Card shuffles: Difference between revisions
m
added a section name to the first talk topic, this will also place the TOC correctly.
(→Overhand shuffle?: new section) |
m (added a section name to the first talk topic, this will also place the TOC correctly.) |
||
| (4 intermediate revisions by 3 users not shown) | |||
Line 1:
== task needs more definition ==
Task needs more definition so we can know whether an implementation is suitable or not. --[[User:Rdm|Rdm]] ([[User talk:Rdm|talk]]) 04:31, 28 April 2015 (UTC)
:It's a draft so OK. What do you want to see? --[[User:Mwn3d|Mwn3d]] ([[User talk:Mwn3d|talk]]) 16:21, 28 April 2015 (UTC)
Line 37 ⟶ 38:
I'm going to implement it "by the book" - which is something of a joke as a shuffle, since it always retains the order of the deck - until we get a better specification. --[[User:Rdm|Rdm]] ([[User talk:Rdm|talk]]) 13:32, 18 May 2015 (UTC)
:The overhand shuffle is implemented in Java as the first few seconds of this video: https://youtu.be/x5tLNHuvf6s. Impossible singlehandedly probably but luckily most people have two hands. The "new" deck is just in the other hand in real life. The way it's explained on Wikipedia does seem a bit off. It should probably say the cuts are taken from the top of the deck and placed on top of the new deck. Neither the riffle nor overhand shuffle are good shuffles in computer science terms. They're just easy for meatbags to do in real life. Don't worry about it actually accomplishing a good shuffle. --[[User:Mwn3d|Mwn3d]] ([[User talk:Mwn3d|talk]]) 03:42, 1 June 2015 (UTC)
== Added definitions ==
I have added definitions of the riffle and overhand shuffles straight on the page so that we can standardize better here. Let me know if anything sounds unclear or incorrect. I'm not rushing to get this task ready for prime time so any corrections can still happen. --[[User:Mwn3d|Mwn3d]] ([[User talk:Mwn3d|talk]]) 14:25, 4 June 2015 (UTC)
== state of the deck before shuffling ==
No mention (unless inferred) was made of the state of the card deck ''before'' shuffling. I had assumed it would be sorted as if the card deck was taken from it's box (from the manufacture), and the cards are always in some sort of order by suits in a very specific order ('''A 2 3 4 5 6 7 8 9 10 J Q K''') for each suit in alternating colors. Normally, one shuffles the ''heck'' out of new deck, and then some standard method is used to shuffle it. Of course, if that was done, there wouldn't be a good way to measure how well (efficient) any one shuffle is after the ''heck'' shuffle, since it is already in a very shuffled state at that point. I had assumed that shuffle effectiveness was a possible goal for this Rosetta Code task. But could it be that the goal is to just program a method of shuffling to compare how to code an algorithm, with a method to force multiple shuffles? So, we're back to a brand new spanking card deck, and a riffle shuffle doesn't do that much, the cards still appear to be in some kind of noticeable and obvious order, but certainly not random. If a single cut would be performed before the shuffle, as is the case for most riffle shuffles, that would help quite a bit. Observing real world riffle shuffles, usually there are: [cut, riffle, ... two more times ... at a minimum]. Some German card games require no more than three riffle shuffles, because three cards are dealt at a time, hoping for clusters of suited cards, thereby increasing the odds of a good bidding hand, but very unlike, say, a bridge game, where the cards are supposed to be very well shuffled. -- [[User:Gerard Schildberger|Gerard Schildberger]] ([[User talk:Gerard Schildberger|talk]]) 06:01, 29 August 2018 (UTC)
Furthermore, most shuffling (in the real world) starts with a barn yard of cards (from playing a hand or game of cards), and the cards are already in some sort of random order as the cards are picked up and pulled together. But of course, that leaves us with the same problem as having a more-or-less random order in the first place, with no obvious way to measure the effectiveness of a particular sort. -- [[User:Gerard Schildberger|Gerard Schildberger]] ([[User talk:Gerard Schildberger|talk]]) 06:01, 29 August 2018 (UTC)
I should mention that a barn yard (or washing machine shuffle) is where all the cards are more-or-less dumped face down in the center of the card table and ... swished around with both hands in some sort of circular pattern, with one hand moving clockwise, the other hand, counter-clockwise, so that most of the cards are interleaved randomly, much like a washing machine might mix cards if it had two agitators (without the water, of course). It has an advantage that many hands can help, but most often, slowing the process a bit. But many hands make light work. -- [[User:Gerard Schildberger|Gerard Schildberger]] ([[User talk:Gerard Schildberger|talk]]) 06:01, 29 August 2018 (UTC)
== riffle shuffle ==
The Rosetta Code task's preamble essentially mentions a perfect riffle shuffle, taking one card from a halved pile, effectively interleaving the 1st card from one pile with the 1st card from the other pile (or the last card ...), and so on with the 2nd two cards, etc. Almost all riffles (in real life) are interleaved in bunches, where a bunch of cards could be one, two, or three, or more, if the cards are well used, and there isn't a good edge for the shuffler's fingers. Using a one, two, or three card
riffle shuffle would be very "real world". Even my mechanical riffle shuffler has a randomizer built-in (caused by unevenness in the playing cards), with the advantage that it doesn't apply hand oils. -- [[User:Gerard Schildberger|Gerard Schildberger]] ([[User talk:Gerard Schildberger|talk]]) 06:01, 29 August 2018 (UTC)
:There are still 2 possibilities. After the perfect split you can perform a perfect riffle such that the previous top card is still on top or is now second. It is well known that if you keep the top card on top each time the deck returns to its original state after 8 iterations.--[[User:Nigel Galloway|Nigel Galloway]] ([[User talk:Nigel Galloway|talk]]) 13:50, 29 August 2018 (UTC)
| |||