Re: secure protocols for mental poker

From: Heiko Stamer (
Date: 07/13/04

Date: Tue, 13 Jul 2004 22:10:06 +0200

On Tue, 06 Jul 2004 20:17:27 -0700, Francesco wrote:

> I'm making an electronic version of a card game for my master
> thesis and I was looking for a good algorithm to deal out cards,

There are some solutions that don't require opening all cards at
the end of the game:

Christian Schindelhauer's "Toolbox for Metal Card Games" [1]
extends ideas of Crepeau's zero-knowledge protocols [2]. However,
the communication complexity depends linearly on number of players
and number of cards.

A recent work is Barnett's and Smart's "Mental Poker Revisited" [3],
who introduced the cryptographic primitive VTMF (verifiable l-out-of-l
threshold masking function) and gave two possible implementations
(discrete logarithm assumption and Paillier's assumption based). Their
protocols will lead to message sizes which are independent of the number
of players and the number of different cards in the game.

Based on [1,3] I'm currently going to write a general (open source)
library for mental card games. Due to lack of time this should not be
finished until December 2004 :-(

[1] Christian Schindelhauer: "Toolbox for Mental Card Games",
    Technical Report A9814, University of Lübeck, 1998

[2] Claude Crepeau: "A zero-knowledge poker protocol that achieves
    confidentiality of player's strategy, or how to achieve an
    electronic poker face", CRYPTO '86, LNCS 263, 1987

[3] Adam Barnett, Nigel P. Smart: "Mental Poker Revisited",
    Cryptography and Coding 2003, LNCS 2898, pp 370--383, 2003

HTH, Regards

Heiko Stamer <>
`Voices` from /usr/src/linux-2.2.19/fs/fat/inode.c:831:
/* this is as close to the truth as we can get ... */