Should I really care if I am using i++, ++i or i += 1 in a loop?

Short answer is that there is no difference. Continue reading for the long answer.

So, again, what is the difference between these statements? Let us use dnSpy and dissassembly to look at the facts.

Starting with i++. In dnSpy this generates following Common Intermediate Language (CIL):

ldloc.0
ldc.i4.1
add
stloc.0

Looking at the dissassembly this gives:

mov         eax,dword ptr [rbp+2Ch] 
inc         eax  
mov         dword ptr [rbp+2Ch],eax 

Onto ++i. In dnSpy this generates following identical CIL:

ldloc.1
ldc.i4.1
add
stloc.1

with identical assembly code:

mov         eax,dword ptr [rbp+28h] 
inc         eax  
mov         dword ptr [rbp+28h],eax 

Same thing for i += 1 - CIL:

ldloc.2
ldc.i4.1
add
stloc.2

Assembly:

mov         eax,dword ptr [rbp+24h] 
inc         eax  
mov         dword ptr [rbp+24h],eax 

So what does this give us in a loop:

for (i = 0; i < 100; i++)
{
  k += i;
}

This results in following assembly (for all 3 cases):

xor         edx,edx
add         eax,edx
inc         edx
cmp         edx,64h
jl          Increment.Program.Main(System.String[]) + 010h(07FFADDF734D0h)

Conclusion: No matter what your mama says, there is no practical difference between i++, ++i or i += 1. Just use the one you like best...