# Why are the FFT outputs always zero for some specific point numbers when using FFTW with MKL

**URL:** <https://discourse.julialang.org/t/why-are-the-fft-outputs-always-zero-for-some-specific-point-numbers-when-using-fftw-with-mkl/137038>\
**Category:** Performance\
**Tags:** question\
**Created:** [May 8, 2026, 12:58pm UTC](https://discourse.julialang.org/t/why-are-the-fft-outputs-always-zero-for-some-specific-point-numbers-when-using-fftw-with-mkl/137038 "2026-05-08T12:58:09Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![DN\_Silence](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/dn_silence/32/218592_2.png) [@DN\_Silence](https://discourse.julialang.org/u/DN_Silence)\
**Post date:** [May 8, 2026, 12:58pm UTC](https://discourse.julialang.org/t/why-are-the-fft-outputs-always-zero-for-some-specific-point-numbers-when-using-fftw-with-mkl/137038/1 "2026-05-08T12:58:10Z")

</div>

I’m using an Intel CPU with Julia 1.12. When my FFT size is 178848185, whether I do FFT or rFFT, the output is always all zeros. But if I use fftw provider, it works normally. This issue is very easy to reproduce. The code is:

using FFTW  
using Random  
FFTW.set\_provider!(“mkl”)  
println(fft(randn(178848185)))

Is this an MKL bug, or are there some usage limitations?

---

<div class="post-metadata">

**Author:** ![giordano](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/giordano/32/2166_2.png) [@giordano](https://discourse.julialang.org/u/giordano)\
**Post date:** [May 8, 2026, 1:44pm UTC](https://discourse.julialang.org/t/why-are-the-fft-outputs-always-zero-for-some-specific-point-numbers-when-using-fftw-with-mkl/137038/2 "2026-05-08T13:44:48Z")

</div>

> [@DN\_Silence](#):
>
> This issue is very easy to reproduce.

Can’t say I can reproduce it

```julia-auto
julia> using FFTW

julia> FFTW.get_provider()
"mkl"

julia> any(iszero, fft(randn(178848185)))
false

```

(`any` looks a bit better than printing 100 million numbers to screen)

> [@DN\_Silence](#):
>
> Is this an MKL bug

Writing a C program doing the same operation and linking to MKL would reveal that.