1 From Andreas Enge 31 August 2011:
2 implement mul_karatsuba with three multiplications at precision around p,
3 instead of two at precision 2*p and one at precision p
4 requires analysis of error propagation
6 From Andreas Enge 30 August 2011:
7 As soon as dependent on mpfr>=3, remove auxiliary functions from
8 get_version.c and update mpc.h.
9 Use MPFR_RND? instead of GMP_RND?, and remove workarounds for MPFR_RNDA from
12 From Andreas Enge 05 July 2012:
13 Add support for rounding mode MPFR_RNDA.
15 From Andreas Enge and Paul Zimmermann 6 July 2012:
16 Improve speed of Im (atan) for x+i*y with small y, for instance by using
17 the Taylor series directly.
20 - from Andreas Enge 9 June 2009:
21 Scripts and web page comparing timings with different systems,
22 as done for mpfr at http://www.mpfr.org/mpfr-2.4.0/timings.html
24 New functions to implement:
25 - from Joseph S. Myers <joseph at codesourcery dot com> 19 Mar 2012: mpc_erf,
26 mpc_erfc, mpc_exp2, mpc_expm1, mpc_log1p, mpc_log2, mpc_lgamma, mpc_tgamma
27 http://lists.gforge.inria.fr/pipermail/mpc-discuss/2012-March/001090.html
28 - from Andreas Enge and Philippe Théveny 17 July 2008
29 agm (and complex logarithm with agm ?)
30 - from Andreas Enge 25 June 2009:
31 correctly rounded roots of unity zeta_n^i
32 - implement a root-finding algorithm using the Durand-Kerner method
33 (cf http://en.wikipedia.org/wiki/Durand%E2%80%93Kerner_method)
34 See also the CEVAL algorithm from Yap and Sagraloff:
35 http://www.mpi-inf.mpg.de/~msagralo/ceval.pdf
38 - from Andreas Enge and Philippe Théveny 9 April 2008
39 correct handling of Nan and infinities in the case of
40 intermediate overflows while the result may fit (we need special code)