discrete fourier transform

[mental sparrow]

first ever c++ project ever. would greatly appreciate it if anyone took a look at the code and perhaps gave me some advice
seeking all advice (optimization, cleaning, best practices etc)

thank you!

https://github.com/checkplease78/dft-impl/blob/main/main.cpp

[jade pilot]

you use exp instead of std::exp

[mental sparrow]

is that an issue? it still produces correct resultss

[jade pilot]

it isn't guaranteed to exist

it's from C's <math.h>which <cmath> includes internally

[mental sparrow]

for

f1 = 3
A2 = 3
f2 = 3```

it produces
```k=2 | f: 2 hz | A: 3 | phi: -1.5708 rad | eqn: A * cos(2*pi*2+-1.5708)
k=3 | f: 3 hz | A: 2 | phi: -1.5708 rad | eqn: A * cos(2*pi*3+-1.5708)```

which is correct

oh gotcha

[jade pilot]

code looks really really clean though

[mental sparrow]

thank you

i feel like it is not optimal though for some reason

[jade pilot]

yes very nice

oh?

[mental sparrow]

all i did to optimize it was precomputing the constants

and compiling using -ffast-math

but not much more

[jade pilot]

hmm

[mental sparrow]

what level is this

of proficiency in c++

is this considered beginner

[jade pilot]

maybe even intermediate, not sure

[pine shoal]

for now until he uses all the other advanced features of c++ to optimize his code

[jade pilot]

like what

[pine shoal]

like concepts, meta programming with templates etc

[jade pilot]

how would those help here

[pine shoal]

so he could check if he using an floating type only or not

etc

[mental sparrow]

is there any room for optimization here?

oh

[pine shoal]

like you can use concepts

like std::floating_point etc

think after you create this library , you tell someone to use it...so instead passing a floating type or double parameters , he passed a int.

so making your own concepts too work