Skip to content

tgamma1pm1 is off by hundreds of epsilon in cpp_bin_float_50 #1519

Closed
#1532
Closed
tgamma1pm1 is off by hundreds of epsilon in cpp_bin_float_50#1519
#1532

Description

@NAThompson
NAThompson
opened on Oct 10, 2026

In cpp_bin_float_50, tgamma1pm1 loses up to about 2,750 epsilon, and the error grows with a; in double it is within half an ulp. The reference is cpp_bin_float_100, checked against Arb to 1e-78.

tgamma1pm1 in cpp_bin_float_50

ligamma inherits this at 50 digits (#1518).

Activity

  1. NAThompson
    closed this as completedin c899254on Oct 10, 2026
  2. JacobHass8
    added a commit that references this issue on Oct 12, 2026

    Evaluate tgamma1pm1 for multiprecision types with a Taylor series (bo…

    c899254
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions