-
When `BigFloat::log` should give `0.5` or `2.0`, it infinite loops (or at least seems to). Precision and rounding mode seem unimportant, but I have been testing with precisions of 64,2048,4096 and `Ro…
mtak- updated
1 month ago
-
Imported from SourceForge on 2024-07-05 23:42:22
Created by **[fateman](https://sourceforge.net/u/fateman/)** on 2023-07-04 17:11:03
Original: https://sourceforge.net/p/maxima/bugs/4152
---
I think …
rtoy updated
2 months ago
-
Imported from SourceForge on 2024-07-09 20:04:25
Created by **[fateman](https://sourceforge.net/u/fateman/)** on 2024-01-03 18:03:48
Original: https://sourceforge.net/p/maxima/bugs/4239
---
airy_dai…
rtoy updated
2 months ago
-
I'm using `astro_float` to validate function approximations in my code. I noticed that one of my tests fails with a massive discrepancy in computed results. Here's a minimal example which reproduces t…
-
With pycftboot, it seems easy to generate XML files which *need* to be processed with pmp2sdp -f json instead of pmp2sdp -f bin. The simplest testcase I could make is attached. In json mode it works p…
-
I have a system where the coefficients of the polynomials are huge integers. When declaring these polynomials, their coefficients are stored as `BigFloat`. When trying to solve the system, the package…
-
```julia
using LinearAlgebra
a = Matrix{BigFloat}(undef, 2,2)
a[1] = 1; a[3] = 1; a[4] = 1
a = Hermitian(a)
b = Symmetric(a)
c = Diagonal([1,1])
d = SymTridiagonal([1 1;1 1])
a+c
a+d
b+c
b+…
-
Is there a way to work with Fourier and BigFloat? It goes via FFTW which is causing some issues.
```
using ClassicalOrthogonalPolynomials
F = Fourier{BigFloat}()
F[:, 1:10] \ sin.(axes(F,1))
``…
-
So this is understandable, and most def not your doing, but currently despite the fact that we can do all of this with BigFloats for crazy accuracy, we can not plot the results. So either mention this…
-
현재 내부적으로는 Int64 자료형을 사용하고 있어서 이 이상으로 가면 오버플로우가 발생합니다.