History of Values of the Barnes $G$-function at rational numbers

back to table · edit · history · where entries came from · files

compare when who what
2026-08-16 17:02 bmatschke how well the digits are known: proven (converted: generators/T93-barnes-g recomputes it with arb (acb_barnes_g); the stored tails were wrong and were corrected) current
2026-08-16 01:02 bmatschke recomputed in ball arithmetic; the last digits were wrong and are now proven reviewed
2026-08-15 11:01 bmatschke how well the digits are known: proven (converted: generators/T93-barnes-g recomputes it with arb (acb_barnes_g); 1080/1080 matched)
2026-08-13 22:07 bmatschke how well the digits are known: heuristic (a fixed-precision value wrapped in an interval field)
2026-08-09 09:16 flattening entries rewritten as records with named parameters
2026-08-09 08:35 data-repository import the current state of the data repository
2021-05-19 19:20 bmatschke from the data repository, e9e464db

What changed between 2026-08-16 01:02 and 2026-08-16 17:02

from line 37 (11 lines, 4 more than before) @@ -37,7 +37,11 @@
   complete: 'no'   rigour: proven-  rigour details: Computed in interval or ball arithmetic throughout, so the digits-    written follow from the width of the result rather than from a guard chosen in-    advance.+  rigour details: 'Recomputed in ball arithmetic (arb, acb_barnes_g), which carries+    its error bound through the computation. This table holds 200 significant digits+    and about 197 of them had been right: mpmath was asked for 200 digits of working+    precision, its barnesg returned fewer correct ones, and nothing checked. Of the+    1050 inexact entries, 995 were outside the one unit in the last place that a written+    value promises -- typically by 10 to 100 units, worst 386. The tails were replaced+    on 2026-08-16; the earlier values are in the table history.' Display properties:   number-header: $G(s)$ 

Sign in to restore an earlier version.