You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I know now consider iPhoneFFT to be irrelevant due to Apple's Accelerator framework, but it looks like there is a slight error in the iPhoneFFT README, which says "Why is the highest frequency computed equal to half the sampling rate?". It looks to me like your code isn't actually including the Nyquist, which according to the Ooura documentation is stored in the 1st element of the output array:
Instead, your code includes the DC component, which like the Nyquist is real-only, and stored in the zero'th element of the output array. It's quite common to discard the DC component and include the Nyquist, but I don't think that's what your code is doing.
Please correct me if I'm wrong, as I'm using Ooura in my own code based on this understanding.
The text was updated successfully, but these errors were encountered:
Should be "right under half the sampling rate".. DC and Nyquist are equally useless in the kind of stuff I was using this codebase for. Thanks for the catch. If it matters to you, can you edit the README and make a pull request?
Best,
Alex
On Jun 19, 2012, at 11:50 AM, Jamie Bullock wrote:
Hi,
I know now consider iPhoneFFT to be irrelevant due to Apple's Accelerator framework, but it looks like there is a slight error in the iPhoneFFT README, which says "Why is the highest frequency computed equal to half the sampling rate?". It looks to me like your code isn't actually including the Nyquist, which according to the Ooura documentation is stored in the 1st element of the output array:
Instead, your code includes the DC component, which like the Nyquist is real-only, and stored in the zero'th element of the output array. It's quite common to discard the DC component and include the Nyquist, but I don't think that's what your code is doing.
Please correct me if I'm wrong, as I'm using Ooura in my own code based on this understanding.
Reply to this email directly or view it on GitHub: #2
Hi,
I know now consider iPhoneFFT to be irrelevant due to Apple's Accelerator framework, but it looks like there is a slight error in the iPhoneFFT README, which says "Why is the highest frequency computed equal to half the sampling rate?". It looks to me like your code isn't actually including the Nyquist, which according to the Ooura documentation is stored in the 1st element of the output array:
Instead, your code includes the DC component, which like the Nyquist is real-only, and stored in the zero'th element of the output array. It's quite common to discard the DC component and include the Nyquist, but I don't think that's what your code is doing.
Please correct me if I'm wrong, as I'm using Ooura in my own code based on this understanding.
The text was updated successfully, but these errors were encountered: