Inaccuracy in symbolic series result on 32bit Linux
From https://trac.sagemath.org/ticket/24838#comment:64
On 32bit:
File "src/sage/functions/hypergeometric.py", line 148, in sage.functions.hypergeometric
Failed example:
hypergeometric_U(2, 2, x).series(x == 3, 100).subs(x=1).n()
Expected:
0.403652637676806
Got:
0.403651580752398
Pynac is not involved in the numerics (hypergeometric_U
is not a GinacFunction
), so the hypothesis is that the expression before .n()
is different with 32bit Linux pynac0.7.22.
Change History (13)
Authors: 
→ Dave Morris

Commit: 
→ e8ee3fc291a8d547459e89ad40f3cf0e7a953d93

Dependencies: 
→ #31585

Keywords: 
pynac 32bit overflow added

Milestone: 
sage8.3 →
sage9.4

Status: 
new →
needs_review

Branch: 
public/25688 →
public/25688r1

Commit: 
e8ee3fc291a8d547459e89ad40f3cf0e7a953d93 →
73e975c28997f6662454b3b53dfe2cdebd22f410

Cc: 
Samuel Lelièvre added

Status: 
needs_review →
needs_work

Branch: 
public/25688r1 →
public/25688r2

Commit: 
73e975c28997f6662454b3b53dfe2cdebd22f410 →
689e08f9393ac651c420546b0e63efbb1220156b

Dependencies: 
#31585

Status: 
needs_work →
needs_review

Milestone: 
sage9.4 →
sage9.5

Cc: 
Thierry Monteil added

Milestone: 
sage9.5 →
sage9.6

Reviewers: 
→ Matthias Koeppe

Status: 
needs_review →
positive_review

Branch: 
public/25688r2 →
689e08f9393ac651c420546b0e63efbb1220156b

Resolution: 
→ fixed

Status: 
positive_review →
closed

This bug is fixed by the pynac patch in #31585. (So the inaccuracy in the sum is due to an arithmetic overflow.)
There is only new commit in the branch here. It just deletes the "known bug" tag from the doctest in src/sage/functions/hypergeometric.py that is in the description of this ticket.
New commits:
trac 31585 pynac overflow patch
add doctests
update pynac patch level
trac 25688 remove "known bug" tag