Given that on the input we discuss that answers returned by msolve don't make sense, we should leave it in place.
IMHO msolve tries to find a univariate representation for the solutions, and so it needs a big field to work in - and I don't even know whether there any pitfalls with this approach for char>0. (or char = 2, often the most pathological case) Dima On 1 July 2024 07:16:21 BST, Marc Mezzarobba <m...@mezzarobba.net> wrote: >Hi Dima, > >Dima Pasechnik wrote: >> So this is due to "0 < char <= 2**17 and deg != elim.degree()" - added >> by you - which does not make sense to me. >> Is this "Criterion" no longer applicable? > >I don't know. This criterion was suggested to me by Mohab after I >complained that msolve -P 1 often returned nonsense in small >characteristic. (As far as I understand, there is always a tiny >probability that the output is incorrect, but at the time and in this >case the probability was not small at all.) I know the developers have >made many improvements since, but I don't know if this particular issue >is solved or not. > -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/sage-devel/EB75AB3D-E94F-438D-B6D3-9CA81D28E759%40gmail.com.
