- 13 Mar, 2010 2 commits
-
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
-
- 12 Mar, 2010 1 commit
-
-
Jean-Marc Valin authored
-
- 01 Mar, 2010 1 commit
-
-
Jean-Marc Valin authored
-
- 26 Feb, 2010 1 commit
-
-
Jean-Marc Valin authored
-
- 12 Feb, 2010 1 commit
-
-
John Ridges authored
-
- 17 Jan, 2010 2 commits
-
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
-
- 16 Jan, 2010 1 commit
-
-
Thorvald Natvig authored
-
- 01 Jan, 2010 1 commit
-
-
Jean-Marc Valin authored
-
- 31 Dec, 2009 1 commit
-
-
Jean-Marc Valin authored
-
- 28 Dec, 2009 2 commits
-
-
Jean-Marc Valin authored
think of) to produce NaNs in the new PLC.
-
Jean-Marc Valin authored
arithmetic.
-
- 21 Dec, 2009 1 commit
-
-
Jean-Marc Valin authored
-
- 17 Dec, 2009 1 commit
-
-
Jean-Marc Valin authored
-
- 15 Dec, 2009 1 commit
-
-
Jean-Marc Valin authored
-
- 11 Dec, 2009 2 commits
-
-
Thorvald Natvig authored
-
Jean-Marc Valin authored
-
- 03 Dec, 2009 2 commits
-
-
Jean-Marc Valin authored
-
George de Vries authored
-
- 02 Dec, 2009 1 commit
-
-
Jean-Marc Valin authored
-
- 26 Nov, 2009 2 commits
-
-
Jean-Marc Valin authored
introduced in Thorvald's previous patch.
-
Jean-Marc Valin authored
-
- 25 Nov, 2009 1 commit
-
-
Thorvald Natvig authored
-
- 22 Nov, 2009 2 commits
-
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
-
- 09 Nov, 2009 1 commit
-
-
Jean-Marc Valin authored
-
- 08 Nov, 2009 1 commit
-
-
Jean-Marc Valin authored
-
- 31 Oct, 2009 2 commits
-
-
Jean-Marc Valin authored
-
Timothy Terriberry authored
Expose the normalized range for reciprocal square roots in fixed-point mode. This allows subsequnt calculations to use the full precision of the result.
-
- 24 Oct, 2009 5 commits
-
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
-
- 23 Oct, 2009 1 commit
-
-
Jean-Marc Valin authored
-
- 22 Oct, 2009 4 commits
-
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
-
Jean-Marc Valin authored
budget at low bitrate.
-
Timothy B. Terriberry authored
Accuracy for rsqrt, rcp, cos, and log2 is now at the level of truncation error for the current output resolution of these functions. sqrt and exp2 still have non-trivial algebraic error, but this cannot be reduced much further using the current method without additional computation. Also updates the fast float approximations for log2 and exp2 with coefficients that give slightly lower maximum relative error. Patch modified by Jean-Marc Valin to leave the cos approximation as is and leave the check for x<-15 in exp2 as is.
-