48K Classic Mode Pitch Algorithm

S2400 Feature Requests
1

Is it possible to reprogram the 48K Classic Mode Pitch Algorithm to sound the same (or as close as possible) as the 26K Pitch Algorithm. I don’t know if it would be simply multiplying the amount of samples dropped, or what would be involved, honestly. But, it’s a shame that the 48K Classic Mode is so subtle it’s almost a waste.

Any thoughts?

I know some will say to just use 26K sampling mode, but I have my reasons for requesting this. And, whether or not they’re willing/able to make such a change dictates whether it’s worth sharing more on the subject.

1 Like
2

Couldn’t you just resample your 48k sample to get the increased aliasing you’re looking for?

3

The classic mode pitch algorithm is the same for 48k mode. you just done hear as much aliasing because nyquest is 24khz vs 26khz where its 13khz. Since there is a lot more audio information at that frequency you will hear the aliasing much more. Using the same pitch algorithm the sample rate makes a huge difference in the grit.

Of course @av500 and @Mickey please feel free to tell me if Im mistaken.

2 Likes
4

Thanks for your reply StupidAmericanPig, I totally understood your reply in my other thread…that’s what lead to me creating this request :+1:

I don’t think the frequency distortion caused by the Pitch Algorithm is technically Aliasing as a lot of folks refer to it. I believe the SP changes pitch by adding/dropping samples, in order to work with the constant master clock. That said, at 26K the added/dropped samples are WAY more noticeable than would be if the same formula was used at 48K, because there’s so many less samples to begin with, at 26K.

I could be totally wrong about everything.

What I’m asking the developers is if it’s possible to make the 48K Classic Mode Pitch Algorithm sound similar to the 26K by increasing the ratio of added/dropped samples at 48K, or some other means. Hopefully one in the know will have the time to chime in.

I don’t use the ESPI app, but one wonders how it maintains the same level of pitch distortion, using any imported sample, at any sample rate. Maybe it resamples to 26K, then back to whatever your computer is set too. I don’t know.

5

Are you testing your 48K samples at 12-bit? That will probably make a difference

6

Yes Sir, makes no difference in pitch artifacts.

7

you cannot make the “classic” pitch algorithm work any different in 48k vs 26k because the algorithm is independent of the sample rate. adding dropping more sample just means changing the pitch down/up more

1 Like
8

Makes sense. Thanks for your response!