| rdfs:comment | - pete-7.c is the largest of the nine entries an anonymous "Pete" submitted to Bignum Bakeoff. It placed third in the competition, behind marxen.c and loader.c. pete-7.c uses a linear array notation defined by the following rules:
* \(f_x(0, 0, \ldots, 0, 0) = x\)
* \(f_x(a_1, a_2, a_3, \ldots, a_{n - 1}, a_n + 1) = f_y(a_1, a_2, a_3, \ldots, a_{n - 1}, 0)\) where \(y = f_x(a_1, a_2, a_3, \ldots, a_{n - 1}, a_n)^2\)
* \(f_x(a_1, a_2, a_3, \ldots, a_n + 1, 0, 0, \ldots, 0, 0) = f_x(a_1, a_2, a_3, \ldots, a_n, x, x, \ldots, x, x)\)
|
| abstract | - pete-7.c is the largest of the nine entries an anonymous "Pete" submitted to Bignum Bakeoff. It placed third in the competition, behind marxen.c and loader.c. pete-7.c uses a linear array notation defined by the following rules:
* \(f_x(0, 0, \ldots, 0, 0) = x\)
* \(f_x(a_1, a_2, a_3, \ldots, a_{n - 1}, a_n + 1) = f_y(a_1, a_2, a_3, \ldots, a_{n - 1}, 0)\) where \(y = f_x(a_1, a_2, a_3, \ldots, a_{n - 1}, a_n)^2\)
* \(f_x(a_1, a_2, a_3, \ldots, a_n + 1, 0, 0, \ldots, 0, 0) = f_x(a_1, a_2, a_3, \ldots, a_n, x, x, \ldots, x, x)\) Define the sequence \(n_0 = 99\) and \(n_{i + 1} = 9 \cdot 2^{n_i}\), letting \(N = n_{16}\). The output value is then \(f_N(N, 0, 0, \ldots, 0, 0)^2\) with \(N - 1\) copies of 0. David Moews, the judge, estimated that pete-7.c is between \(f_{\omega^\omega}(2\uparrow\uparrow 35)\) and \(f_{\omega^\omega}(2 \uparrow\uparrow 36)\) in the fast-growing hierarchy.
|
