diff options
author | Márton Danóczy <marton.danoczy@nielsen.com> | 2015-09-30 18:12:31 +0200 |
---|---|---|
committer | Márton Danóczy <marton.danoczy@nielsen.com> | 2015-09-30 18:12:31 +0200 |
commit | 64b8a2ed6a1b624849e7b369959524d346c598d9 (patch) | |
tree | 30af23aee3463db8b688b18e09fe0f1f5cfb8d16 | |
parent | 16490be31ba96c11eae52bde754351c05f604cda (diff) | |
download | pffft-64b8a2ed6a1b624849e7b369959524d346c598d9.tar.gz |
readme in markdown
-rw-r--r-- | README.md (renamed from README.txt) | 17 |
1 files changed, 6 insertions, 11 deletions
@@ -1,7 +1,6 @@ -PFFFT: a pretty fast FFT. +# PFFFT: a pretty fast FFT. -TL;DR --- +## TL;DR PFFFT does 1D Fast Fourier Transforms, of single precision real and complex vectors. It tries do it fast, it tries to be correct, and it @@ -10,8 +9,7 @@ on x86 cpus, Altivec on powerpc cpus, and NEON on ARM cpus. The license is BSD-like. -Why does it exist: --- +## Why does it exist: I was in search of a good performing FFT library , preferably very small and with a very liberal license. @@ -63,15 +61,13 @@ thought, especially with the real transforms, and I ended up writing more code than I planned.. -The code: --- +## The code: Only two files, in good old C, pffft.c and pffft.h . The API is very very simple, just make sure that you read the comments in pffft.h. -Comparison with other FFTs: --- +## Comparison with other FFTs: The idea was not to break speed records, but to get a decently fast fft that is at least 50% as fast as the fastest FFT -- especially on @@ -88,8 +84,7 @@ provides "unordered" FFTs , and a fourier domain convolution operation. -Benchmark results (cpu tested: core i7 2600, core 2 quad, core 1 duo, atom N270, cortex-A9) --- +## Benchmark results (cpu tested: core i7 2600, core 2 quad, core 1 duo, atom N270, cortex-A9) The benchmark shows the performance of various fft implementations measured in MFlops, with the number of floating point operations being defined as 5Nlog2(N) |