Pascal
|
50548c194e
|
doc
|
2017-10-30 08:43:00 +01:00 |
|
Pascal
|
c4b2b0968f
|
debug messages
|
2017-10-30 08:43:00 +01:00 |
|
Pascal
|
46cceb956a
|
use dft convol for smooth integers
|
2017-10-30 08:43:00 +01:00 |
|
Pascal
|
4d880302bd
|
catch power of 2 in convol
|
2017-10-30 08:43:00 +01:00 |
|
Pascal
|
3d4c497c32
|
allow input aliasing + input shift in rad2 and bluestein
|
2017-10-30 08:43:00 +01:00 |
|
Pascal
|
104ed478f8
|
detect rad2
|
2017-10-30 08:43:00 +01:00 |
|
Pascal
|
04b5dcf3a6
|
document acb_dft
|
2017-10-30 08:43:00 +01:00 |
|
Fredrik Johansson
|
a242af9aab
|
quick fix in sinh_cosh to get smaller enclosure for small midpoint + large radius; plus faster handling of inf/nan
|
2017-10-28 21:04:42 +02:00 |
|
Fredrik Johansson
|
6599fb8528
|
use interval Newton update in legendre_p_ui_root
|
2017-10-25 18:23:53 +02:00 |
|
Fredrik Johansson
|
ac044b4e29
|
slight speedup of Legendre polynomials
|
2017-10-18 15:17:52 +02:00 |
|
Fredrik Johansson
|
0346fe4df7
|
slight tuning of cutoff parameters
|
2017-10-13 23:57:55 +02:00 |
|
Fredrik Johansson
|
0021661d61
|
faster basecase for Legendre polynomials, and a bugfix
|
2017-10-13 22:09:19 +02:00 |
|
Fredrik Johansson
|
1f9242d626
|
use old flint function
|
2017-10-11 02:03:14 +02:00 |
|
Fredrik Johansson
|
61f9e6c489
|
computation of Legendre polynomial roots and Gauss-Legendre quadrature nodes
|
2017-10-11 00:37:57 +02:00 |
|
Fredrik Johansson
|
8947049b77
|
fix and refactor derivative bounds for legendre polynomials
|
2017-10-10 22:34:09 +02:00 |
|
Fredrik Johansson
|
49e2d28582
|
fix severe error overestimate in legendre_p_ui
|
2017-10-10 16:04:27 +02:00 |
|
Fredrik Johansson
|
ffaed26841
|
fast and accurate evaluation of ordinary Legendre polynomials (may need further work)
|
2017-10-09 21:29:48 +02:00 |
|
Fredrik Johansson
|
df40b73eee
|
doc fix
|
2017-10-06 20:08:15 +02:00 |
|
Fredrik Johansson
|
44bb510943
|
add mag_bin_uiui
|
2017-10-06 20:07:41 +02:00 |
|
Fredrik Johansson
|
ebfe2189b9
|
asymptotic expansion of Legendre polynomials, initial code
|
2017-10-05 16:29:16 +02:00 |
|
Pascal
|
33dafb4ff5
|
add acb_dft function
|
2017-10-05 13:37:35 +02:00 |
|
Fredrik Johansson
|
76ffc69264
|
fast computation of central binomial coefficients (todo: improve general bin_uiui)
|
2017-10-04 22:01:40 +02:00 |
|
Pascal
|
8e13588eb4
|
update dirichlet profile code
|
2017-10-04 21:25:30 +02:00 |
|
Pascal
|
b96ae03f7a
|
enable bluestein, need to find cutoff pt
|
2017-10-04 16:15:58 +02:00 |
|
Pascal
|
842e334367
|
fix acb_dirichlet_dft test -- the reference was wrong
|
2017-10-04 15:46:45 +02:00 |
|
Pascal
|
fee1379331
|
fix bluestein fft
|
2017-10-02 18:08:55 +02:00 |
|
Pascal
|
cfa3545e2b
|
simplify paths
|
2017-10-02 14:51:23 +02:00 |
|
Fredrik Johansson
|
28c969927c
|
merge and update some acb_dirichlet code
|
2017-09-18 18:20:47 +02:00 |
|
Fredrik Johansson
|
aa30e792ae
|
acb_atan: improve error bounds
|
2017-09-18 00:06:26 +02:00 |
|
Fredrik Johansson
|
b84da69711
|
fix a typo
|
2017-09-11 04:14:35 +02:00 |
|
Fredrik Johansson
|
d3e395f4f4
|
speed up arb_exp() between 4600 and 19000 bits
|
2017-09-06 16:45:56 +02:00 |
|
Fredrik Johansson
|
fcd3d3bf12
|
sin, cos: replace mpfr with faster custom implementation of the bit-burst algorithm at very high precision
|
2017-09-04 04:43:07 +02:00 |
|
Fredrik Johansson
|
82fff1db45
|
use generic arb version of rs algorithm instead of mpfr to speed up sin/cos at 4600 to 90000 bits; also always use arb's pi for argument reduction
|
2017-09-03 01:30:13 +02:00 |
|
Fredrik Johansson
|
acc552d2ee
|
fix a doc typo
|
2017-08-25 19:17:56 +02:00 |
|
Fredrik Johansson
|
380ce57f6a
|
reduce z' mod tau' in theta functions to support larger arguments with reasonable performance
|
2017-08-25 15:40:57 +02:00 |
|
Fredrik Johansson
|
64fd8a8920
|
speed up acb_elliptic_rj for high precision by using a variable length Taylor expansion
|
2017-08-23 22:51:20 +02:00 |
|
Fredrik Johansson
|
a275e740b3
|
speed up acb_elliptic_rf slightly by precomputing optimal denominators
|
2017-08-21 03:07:24 +02:00 |
|
Fredrik Johansson
|
2fb86a0383
|
speed up acb_elliptic_rf for high precision by using a variable length Taylor expansion
|
2017-08-17 23:51:03 +02:00 |
|
Fredrik Johansson
|
ac713b31aa
|
speed up acb_agm1 by finishing with Taylor series
|
2017-08-12 21:05:54 +02:00 |
|
Fredrik Johansson
|
757d714ac1
|
speed up arb_agm
|
2017-08-12 15:03:09 +02:00 |
|
Fredrik Johansson
|
2ea1d58373
|
arb_rel_accuracy_bits: add fast path
|
2017-08-12 00:48:31 +02:00 |
|
Fredrik Johansson
|
8b9e60cae2
|
remove duplicated macro
|
2017-08-12 00:47:51 +02:00 |
|
Fredrik Johansson
|
46bfe1e976
|
add mag_set_d_lower, mag_sqrt_lower, mag_set_d_2exp_fmpz_lower; some code improvements
|
2017-08-11 11:33:20 +02:00 |
|
Fredrik Johansson
|
bd243ec590
|
arf_set_d, fmpr_set_d: fix handling of subnormals
|
2017-08-11 02:26:16 +02:00 |
|
Fredrik Johansson
|
683dea6c17
|
agm: use central difference instead of forward difference to reduce precision
|
2017-08-08 23:36:56 +02:00 |
|
Fredrik Johansson
|
6035ee2420
|
add acb_mat_mul_threaded
|
2017-08-01 14:01:05 +02:00 |
|
Fredrik Johansson
|
de4fc234ea
|
update version to 2.11.1
|
2017-07-10 19:23:44 +02:00 |
|
Fredrik Johansson
|
656ec293c2
|
avoid n_mulmod2 also in the test code
|
2017-07-10 18:24:14 +02:00 |
|
Fredrik Johansson
|
1d80b22fb9
|
avoid n_mulmod2 which is not available in the released flint
|
2017-07-10 17:46:09 +02:00 |
|
Fredrik Johansson
|
b8b0f47d13
|
fix .build_dependencies to test flint 2.5 instead of flint trunk
|
2017-07-10 17:43:56 +02:00 |
|