Zulip Chat Archive
Stream: mathlib4
Topic: field_simp made no progress
Asei Inoue (Jul 03 2024 at 11:33):
why this raise an error?
import Mathlib.Tactic
variable (x y : ℝ)
example : x * y ≤ (x ^ 2 + y ^ 2) / 2 := by
/-
simp made no progress
-/
field_simp
Damiano Testa (Jul 03 2024 at 11:43):
To me, it does seem that (field_)simp
made no progress, so I don't find the error strange.
Asei Inoue (Jul 03 2024 at 11:44):
my expectation: goal is changed to 2 * x * y ≤ (x ^ 2 + y ^ 2)
actual: no progress
Damiano Testa (Jul 03 2024 at 11:52):
I have nothing else but this example to go by, but it seems that "your" normal form is not one chosen by field_simp
. I know that there has been talk about re-vamping field_simp
, but I do not know what the, possibly future, normal form for this expression would be.
Ruben Van de Velde (Jul 03 2024 at 11:59):
This seems like a logical expectation. Does it work with equality?
Michael Stoll (Jul 03 2024 at 12:22):
Yes. I think inequalities are out of scope for field_simp
.
Michael Stoll (Jul 03 2024 at 12:23):
(They require some order structure in addition, so cannot be dealt with using only field properties.)
Asei Inoue (Jul 03 2024 at 12:36):
indeed.
Asei Inoue (Jul 03 2024 at 12:36):
mere field does not have order….
Asei Inoue (Jul 03 2024 at 14:54):
is there a tactic can handle order relation in field?
Kevin Buzzard (Jul 03 2024 at 17:11):
nlinarith
Last updated: May 02 2025 at 03:31 UTC